This item is only available as the following downloads:
i Bdippza : Synthesis and Metal Complexes of a N ew Monoanionic [ N 2 O] Heteroscorpionate Ligand By Benjamin M. Kriegel A Thesis Submitted to the Division of Natural Sciences New College of Florida In partial fulfillment of the requirements for the degree of Bachelor of Arts in Chemistry Under the sponsorship of Dr. Suzanne E. Sherman Sarasota, Florida May, 2010
ii Acknowledgments First and foremost, I would like to thank Dr. Suzanne Sherman for sponsoring my thesis, allowing me to carry out my thesis resea rch in her laboratory, and providing me valuable support and advice throughout the year. Second, I would like to thank Dr. Steven Shipman for the valuable experience I had working in his lab over summer 2009 and January 2010, for being supportive of my de cision to pursue inorganic chemistry, and for the substantial amount of both scientific and life advice he has given me throughout the last two years. Third, I would like to thank Dr. Paul Scudder for all of his practical advice in organic synthesis and N MR s pectroscopy throughout the year and for giving me the tools to understand organic chemistry and develop a chemical intuition. I thank Dr. Lee Daniels at Rigaku for X ray crystallographic structure determinations; Erinn Brigham and Eric Andreansky for helping me learn various lab skills and helping to acclimate me to the lab as well as for helpful advice ; Kaitlin Lovering for all of her ideas that went into this project, and Patrick McCarthy for helping with various issues whenever they arose I also thank Tricia Chua for all the time she spent in the lab with me so that I could come in at odd times, and for being pretty much totally awesome I acknowledge Dr. Gerard Parkin and his group, particularly Aaron Sattler, Wes Sattler, and Ahmed Al Harbi for teaching me many of the techniques I use in the lab, and for giving me my first exposure to laboratory research in inorganic synthesis. This experience led me to the decision to pursue inorganic chemistry in my graduate research. Finally, I would like to thank New College Foundation, the Alumni Association and the Council of Academic Affairs for funding.
iii Abstract The tridentate heteroscorpionate ligand bis 3,5 diisopropylpyrazol 1 ylacetate (bdippza) has been prepared via a simple three step synthesis, and some zinc and magnesium complexes containing this ligand have been characterized via 1 H NMR, 13 C NMR, and FT IR spectroscopy and by single crystal X ray diffraction. The bis ligand complex Zn(bdippza) 2 has been isolated from the reaction of bdippza and zinc triflate, and structurally characterized via single crystal X ray diffraction The analogous magnesium complex, Mg(bdippza) 2 has also been isolated. The carboxylate bridged dimer [Zn(bdippza)Cl] 2 has also been isolated from the reaction of bdippza a nd zinc chloride. The relevance of bdippza complexes to synthetic structural and functional modeling of the active sites of metalloenzymes, such as the ubiquitous enzyme RuBisCo is also discussed.
iv Table of Contents 1 Introduction and Applications of S corpionate Ligands in Bioi norganic Chemistry 1.1 Tp and O ther Boron Centered Scorpionate Ligands 1 1.2 Carbo n Centered Scorpionate Ligands 17 1.3 Bis Pyrazolylacetates and Related [N 2 O] Scorpionate Ligands 23 1.4 Bdippza: A new ligand for modeling RuBisCo 37 2 Experimental 2.1 General Considerations 47 2.2 Preparation of Ligands 48 2.3 Preparation of Metal Complexes 51 3 Results and Discussion 3.1 Ligand Synthesis: Analysis of Reaction Mechanisms and Spectra 55 3.2 Metal Complexes: Synthesis, Spectra, and Crystallographic Data 7 1 4 C onclusions and Future Directions 90 A Appendices A .1 Appendix A : Selected NMR and FT IR Spectra 96 A .2 Appendix B : Crystallographic Information 112 Works Cited 13 6
v List of Figures Figure 1: Tp 3 [N 3 ] coordination to a metal center ....................... 2 Figure 2: One of seve ral equivalent resonance structures of a Cp donor (left) and a Tp donor (right) to a metal center ................................ ................................ ............................ 3 Figure 3: Tp 2 Zn, one example of a bis ligand complex generated from Tp ligands containing little steric hindrance 1 ................................ ................................ ........................ 4 Figure 4: Tp tBu ZnCl, one example of a distorted tetrahedral complex generated from bulky Tp ligands 9 ................................ ................................ ................................ ................ 4 Figure 5: Tp tBu 2 NO 2 ), an example of a 5 coordinate complex containing a sterically hindered Tp tBu liga nd 8 ................................ ................................ ................................ ......... 4 Figure 6: First row Tp iPr,4Br transition metal complexes exhibit [N 3 ] coordination with the unhindered ligand Tp (left), and [N 2 H] coordination with the more hindered Tp Ph (right) 10,11 (Adapted from ref. 2) ................................ ................................ ........................ 5 Figure 7: An example of a sterically hindered Tp iPr2 Fe complex exhibiting a 6 coordinate structure, in which a molecule of acetonitrile solvent and a molecule of benzoate are coordinated in addition to Tp iPr2 (Taken from ref. 2) ................................ .......................... 6 Figure 8: Representation of the active site of the enzyme carbonic anhydrase in its resting state (Taken from ref. 17) ................................ ................................ ................................ ... 8 Figure 9: Tp tBu,Me ZnOH, the first structural model for the active site of carbonic anhydrase (Taken from ref. 17) ................................ ................................ .......................... 8 Figure 10: Solution phase structure of Tp tBu,Me ZnOCOOH prepared from addition of CO 2 to Tp tBu,Me ZnOH 22 (Taken from ref. 17) ................................ ................................ ........... 11 Figure 11: Solid state structure of Tp tBu,Me ZnOCOOZn Tp tBu,Me isolated from reaction of CO 2 with Tp tBu,Me ZnOH (Taken from ref. 17) ................................ ................................ .. 11 Figure 12: Solid state structure of Tp tBu,Me ZnOCOMe isolated from reaction of dimethylpyrocarbonate and Tp tBu,Me ZnOH (Taken from ref. 17) ................................ ..... 11 Figure 13: An [S 3 ] monoanionic hydrotris mercaptoimidazolylborate (Tm R ) ligand ...... 15 Figure 14: A heteroscorpionate zinc complex containing two mercaptoimidazol donors and one pyrazole donor used to model liv er alcohol dehydrogenase (LADH) (Taken from ref. 29) ................................ ................................ ................................ ............................... 16 Figure 15: Hydrotris(selenoimidazolyl)borate (Tse R ) and Hydrotris(oxoimidazolyl)borate (To R ), recently published [Se 3 ] and [O 3 ] homoscorpionate ligands analogous to the more commonly used Tm R ................................ ................................ ................................ ......... 16 Figure 16: A tris(pyrazolyl)methane (Tp ) ligand, a neutral analog to the monoanionic Tp ligand ................................ ................................ ................................ ...................... 17 Figure 17: ORTEP diagram of tpms tBu ZnBr, in which the ligand is bound [N 2 O] r ather than [N 3 ]. (Taken from ref. 43) ................................ ................................ ........................ 22 Figure 18: ORTEP diagram of acetato bridged dimer prepared from tpms tBu ZnEt and 0.5 equivalents acetic acid. (Taken from ref. 43) ................................ ................................ .. 23
vi Figure 19: Bis[(3,5 diisopro pylpyrazolyl)ethyl]ether is an [N 2 O] ligand that binds meridonally, rather than facially, to zinc. (Taken from ref. 45) ................................ ...... 24 Figure 20: (3 Tert butyl 2 hydroxy 5 methylphenyl) bis(3,5 dimethylpyrazol 1 yl)methane binds as an [N 2 O] facial ligand to produce tetrahedral zinc complexes. (Taken from ref. 17) ................................ ................................ ................................ ...................... 25 Figure 21: ORTEP diagram of Zn(bdmpza) 2 in which two equivalents of ligand bind to an octahedral zinc center. (Taken from ref. 44) ................................ ............................... 28 Figure 22: ORTEP diagram of Bpa tBu2,Me2 ZnMe, a chiral, tetrahedral bis(pyrazolyl)acetate complex. (Taken from ref. 44) ................................ ...................... 29 Figure 23: ORTEP diagram of Bpa tBu2,Me2 ZnOH 2 + which is a carboxylate bridged polymer in the solid state. (Taken from ref. 44) ................................ .............................. 30 Figure 24: The 5 coordinate iron compound isolated from reaction of [Fe(bdtbpza)Cl] 2 with benzoylformate is a good structural model for iron oxygenase enzymes bound to a succinate molecule. (Taken from ref. 44) ................................ ................................ ........ 37 Figure 25: TACN monoacetate magnesium complex prepared by the Sherman group in initial attempts to model RuBisCo. (Taken from ref. 57) ................................ ................ 40 Figure 26: Acetol can be used as a low molecula r weight analog of D ribulose 1,5 bisphosphate, the substrate for RuBisCo, in synthetic model studies. The atoms labeled with the same numbers are analogous to one another. ................................ ...................... 41 Figure 27: The proposed ligand, bdippza, is intermedi ate in sterics between bdmpza and bdtbpza ................................ ................................ ................................ .............................. 42 Figure 28: A goal of this project is to prepare tetrahedral zinc complexes of the structure shown ................................ ................................ ................................ ................................ 44 Figure 29: 1 H NMR spectrum of bdmpzm with DEBA impurity in CDCl 3 Assignments of peaks are labeled. The unlabeled peaks correspond to TMS, vacuum grease, H 2 O, and CHCl 3 going from right to left. ................................ ................................ ........................ 63 Figure 30: 1 H NMR spectrum of Hbdippza in CDCl 3 Assign ments of peaks are labeled. ................................ ................................ ................................ ................................ ........... 68 Figure 31: 13 C NMR spectrum of Hbdippza in CDCl 3 Assignments of peaks are labeled. ................................ ................................ ................................ ................................ ........... 70 Figure 32: 1 H NMR spectrum of Zn(bdippza) 2 in CDCl 3 ................................ ................ 73 Figure 33: ORTEP diagram of Zn(bdippza) 2 Atoms in the asymmetric unit are labeled. ................................ ................................ ................................ ................................ ........... 75 Figure 34: 1 H NMR spectrum of the product isolated from the reaction of bdippza and ZnCl 2 taken in CDCl 3 ................................ ................................ ................................ ....... 81 Figure 35: Preliminary x ray crystallographically determined structure of [Zn(bdippza)Cl] 2 Atoms are represented as spheres (grey carbon, blue nitrogen, red oxygen, maroon zinc, yellow chlorine). Hydrogen atoms are omitted for clarity. 82 Figure 36: Structure of Zn(bpa tBu2,Me2 ) 2 reported by Burzlaff (Taken from ref. 3). ......... 83 Figure 37: 1 H NMR spectrum of Mg(bdippza) 2 in CDCl 3 ................................ ................ 85
vii Figure 38: 1 H NMR of Na(bdippza) in CD 3 OD ................................ ................................ 87 Figure 39: 1 H NMR of mixture of bdippza and Mg(OTf) 2 in CD 3 OD ............................. 88 Figure 40: 1 H NMR spectrum of Mg(bdippza) 2 and Mg( n Bu) 2 in CDCl 3 Peaks corresponding to the butyl protons overlap with the methyl peaks of M g(bdippza) 2 between 0.9 and 1.6 ppm. The peak at 3.65 ppm likely corresponds to THF solvent. .... 90 Figure 41: 1 H NMR spectrum of 3,5 diisopropylpyrazole in CDCl 3 ................................ 96 Figure 42: 13 C NMR spectrum of 3,5 diisop ropylpyrazole in CDCl 3 .............................. 97 Figure 43: 1 H NMR spectrum of bdippzm in CDCl 3 ................................ ........................ 98 Figure 44: 13 C NMR spectrum of bdippzm in CDCl 3 ................................ ....................... 99 Figure 45: 1 H NMR spectrum of Hbdippza in CDCl 3 ................................ .................... 100 Figure 46: 13 C NMR spectrum of Hbdippza in CDCl 3 ................................ ................... 101 Figure 47: Solid state FT IR spectrum of Hbdippza ................................ ...................... 102 Figure 48: 1 H NMR spectrum of Zn(bdippza) 2 in CDCl 3 ................................ ............... 103 Figure 49: 13 C NMR spectrum of Zn(bdippza) 2 in CDCl 3 ................................ .............. 104 Figure 50: Solid state FT IR spectrum of Zn(bdippza) 2 ................................ ................. 105 Figure 51: 1 H NMR spectrum of [Zn(bdippza)Cl] 2 in CDCl 3 ................................ ........ 106 Figure 52: 13 C NMR spectrum of [Zn(bdippza)Cl] 2 in CDCl 3 ................................ ....... 107 Figure 53: Solid state FT IR spectrum of [Zn(bdippza)Cl] 2 ................................ .......... 108 Figure 54: 1 H NMR spectrum of Mg(bdippza) 2 in CDCl 3 ................................ .............. 109 Figure 55: 13 C NMR spectrum of Mg(bdippza) 2 in CDCl 3 ................................ ............. 110 Figure 56: Solid state FT IR spectrum of Mg(bdippza) 2 ................................ ................ 111 List of Schemes Scheme 1: Proposed mechanism for conversion of CO 2 to HCO 3 with carbonic anhydrase (Taken from ref. 17) ................................ ................................ ........................ 10 Scheme 2: In contrast to the reversible reaction of Tp tBu,Me ZnOH with CO 2 Tp iPr2 ZnOH undergoes irreversible dimer formation (Taken from ref. 17) ................................ .......... 12 Scheme 3: Tp tBu,Me ZnOH 2 + a model for protonated CA, was first prepared by reaction with (C 6 F 5 ) 3 B OH 2 in Et 3 N (Taken from ref. 17) ................................ ............................. 13 Scheme 4: General synthesis for poly azolylmethanes proposed by Elguero and coworkers (Adapted from ref. 39) ................................ ................................ .................... 18 Scheme 5: Tpzm ligands can be readily functionalized by deprotonation of the central carbon with a strong base followed by reaction with an electrophilic substrate; in this case, formaldehyde. (Taken from ref. 36) ................................ ................................ ........ 19 Scheme 6: Several cadmium coordination polymers can be prepared using tpzm ligands linked through a phenyl group. (Adapted from ref. 41) ................................ ................... 20
viii Scheme 7: The water soluble tris pyrazolylmethanesulfonate ligand can be pre pared from treatment with sulfur trioxide after deprotonation of the central carbon. (Taken from ref. 42) ................................ ................................ ................................ ................................ ..... 21 Scheme 8: An equilibrium exists between [N 3 ] and [N 2 O] tpms tBu ZnX (X = Cl, Br, Et) in solution. (Taken from ref. 43) ................................ ................................ .......................... 22 Scheme 9: The [N 2 O] facial binding scheme common in enzyme active sites was first accessed by inserting carbon dioxide or ketones/aldehydes into the B H Zn bond in Bp complexes. (Taken from ref. 17) ................................ ................................ ...................... 25 Scheme 10: Two step synthesis of Hbdtbpza in which bdtbpzm is prepared, and then functionalized with CO 2 (Taken from ref. 44) ................................ ................................ .. 27 Scheme 11: Alternative synthesis of Hbdmpza (R = CH 3 ) and Hbpza (R = H). The reaction generally resu step. (Taken from ref. 44) ................................ ................................ ................................ 27 Scheme 12: Reaction of bdtbpza with zinc chloride results in a tetrahedral zinc complex containing one equivalent of ligand. (Taken from ref. 44) ................................ .............. 28 Scheme 13: Preparation of (bdtbpza)ZnSCH 2 Ph by reaction of (bdtbpza)ZnCl with HSCH 2 Ph under basic conditions (Taken from ref. 44) ................................ .................. 31 Scheme 14: 5 Coordinate zinc complexes that model in hibitors bound to proteases have been prepared from (bdtbpza)ZnMe. (Taken from ref. 44) ................................ ............. 32 Scheme 15: Many iron oxidases catalyze hydroxylation of a substrate, coupled to conversion of 2 oxoglutarate to succinate and CO 2 (Taken from ref. 44) ...................... 33 Scheme 16: General mechanism for 2 oxoglutarate dependent iron oxygenase enzymes. (Taken from ref. 44) ................................ ................................ ................................ .......... 34 Scheme 17 : Reaction of bdtbpza with ferrous chloride results in a carboxylate bridged dimer in the solid state. (Taken from ref. 44) ................................ ................................ .. 36 Scheme 18: Proposed reaction mechanism of RuBisCo. M represents an Mg 2+ center. (Taken from ref. 56 ) ................................ ................................ ................................ .......... 39 Scheme 19: 2,4 Di tert butyl 6 (bis(3,5 dimethylpyrazolyl)methyl)phenol reacts with magnesium and sodium sources to form a variety of tetrahedral and octahedral compounds. (Taken from ref. 60) ................................ ................................ .................... 45 Sc heme 20: An objective of this project is to prepare mononuclear zinc and magnesium complexes in which a molecule of acetol can coordinate, in order to potentially model the enediolization step carried out at the active site of RuBisCo. ................................ .......... 47 Scheme 21: Preparation of 3,5 diisopropylpyrazole from 1,6 dimethyl 3,5 heptanedione and hydrazine hydrate ................................ ................................ ................................ ....... 57 Scheme 22: Proposed reaction mechanism for conversion of 1,6 dimethyl 3,5 heptanedione to 3,5 diisopropylpyrazo le. Proton transfer steps are omitted. ................... 58 Scheme 23: Preparation of bdippzm from 3,5 diisopropylpyrazole and dichloromethane ................................ ................................ ................................ ................................ ........... 60
ix Scheme 24: Proposed reaction mechanism for conversion of 3,5 diisopropylpyra zole to bdippzm ................................ ................................ ................................ ............................. 61 Scheme 25: A) S N 2 mechanism for conversion of TEBA to DEBA B) E2 mechanism for conversion of TEBA to DEBA ................................ ................................ ......................... 64 Scheme 26: Preparation of Hbdippza from bdmpza and carbon dioxide ......................... 65 Scheme 27: Proposed reaction mechanism for conversion of bdippzm to bdippza ......... 67 Scheme 28: Preparation of Zn(bdippza) 2 from Hbdippza and zinc triflate in basic methanol ................................ ................................ ................................ ............................ 71 Scheme 29: P reparation of [Zn(bdippza)Cl] 2 from Hbdippza and zinc chloride in basic methanol ................................ ................................ ................................ ............................ 79 Scheme 30: Preparation of Mg(bdippza) 2 from Hbdippza and MgX 2 in basic methanol, where X = OTf or Cl ................................ ................................ ................................ ......... 83 Scheme 31: Propo sed synthesis of Hbotbia, a compound that may act as an [O 3 ] heteroscorpionate ligand in deprotonated form ................................ ................................ 93 List of Tables Table 1: Data acquisition and structure determination details for Zn(bdippza) 2 .............. 76 Table 2 : Selected bond distances and angles of Zn(bdippza) 2 shown in comparison to selected bond distances and angles of Zn(bdmpza) 2 obtained from ref. 51 ...................... 77 Table 3: Dihedral angles of methyl groups with respect to the pyrazolyl planes ............. 78 Table 4: Distances between O2 and hydrogen atoms from another molecule of Zn(bdippza) 2 within the crystal lattice. Hydrogen atoms are labeled with the number of the carbon atom to which they are bonded. ................................ ................................ ...... 79
x List of Abbreviations ACE angiotensin converting enzyme bdmpza bis 3,5 dimethylpyrazol 1 ylacetate bdmpzm bis 3,5 dimet hylpyrazolylmethane bdippza bis 3,5 diisopropylpyrazolyl 1 acetate bdippzm bis 3,5 diisopropylpyrazolylmethane bdtbpza bis 3,5 di tert butylpyrazol 1 ylacetate bdtbpzm bis 3,5 di tert butylpyrazol 1 ylmethane Bpa tBu2,Me2 (3,5 di tert butylpyrazol 1 yl)(3 dimethylpyrazol 1 yl)acetate bpza bis pyrazol 1 ylacetate botbia bis 2 oxo 1 tert butylbenzimidazolyl acetate botbim bis 2 oxo 1 tert butylbenzimidazol 3 ylmethane CA carbonic anhydrase CPA carboxypeptidase A CPB carboxypeptidase B DEBA diethyl benzylamine RuBisCo ribulose 1,5 bisphosphate carboxylase
xi TACN 1,4,7 triazacyclononane tdmpzm tris 3,5 dimethylpyrazolylmethane tdtbpzm tris 3,5 di tert butylpyrazolylmethane TEBA triethylbenzylammonium chloride TLN thermolysin Tm Hydrotris(mercaptoimi dazolyl)borate To Hydrotris(oxoimidazolyl)borate Tse Hydrotris(selenoimidazolyl)borate Tp hydrotrispyrazolylborate Tp* hydrotris 3,5 dimethylpyrazolylborate tpms tBu tris 3 tert butylpyrazolylmethanesulfonate tpzm tris pyrazolylmethane ZBG zinc bind ing group
1 Chapter 1: Introduction and Applications of Scorpionate Ligands in Bioi norganic Chemistry 1.1 Tp and Other Boron Centered Scorpionate Ligands The development of new multidentate ligands that can be used to tailor the structural and electronic properties at a metal center has been an active area of research in inorganic chemistry throughout the last century. Due to the chelate effect, ligands that bind at multiple coordination sites often bind irreversibly to a metal center, influencing chemist ry at other coordination sites while remaining tightly bound, thus making them useful in controlling the environment at a metal to make complexes with specific reactivity. In 1967, Swiatoslaw Trofimenko, a research chemist at DuPont, reported the synthesis of the first of a new class of tridentate ligand, h ydrotrispyrazolylborate (Tp) shown in Figure 1 1 scribe this and similarly structured ligan ds, creating an analogy between the binding do pyrazoles and the claws of a scorpion and the third pyrazole to 2 The similarities between Tp and a scorpion are apparent upon studying the structure of Tp metal complexes. Specifically, the structure of the ligand requires that a six membered chelate ring with a boat structure is present when the Loosely, the term scorpionate is now used to refer to any tridentate ligand that is centered at a single atom with at least two identical substituents, each of which binds to a metal one to three (but usually t wo) atoms away from the central atom. Although not technically fitting the formal definition, chiral ligands containing three distinct
2 substituents that all bind to one metal center are also often referred to as scorpion ate ligands. 3 4 5 refers to ligands in which the sting is identical to s to ligands in which the sting and claws are distinct from one another. Likely due to the relative ease of their synthesis, homoscorpionates have been much more extensively studied than heter oscorpionates in the literature, although more focus has been given to heteroscorpionates in recent years. 2 6 Figure 1 : Tp exhibiting s 3 [N 3 ] coordination to a metal center Tp ligands are analogous in many respects to c yclopentadienyl (Cp) ligands which are among the most extensively used ligands in inorganic chemistry Both have identical symmetry properties in metal complexe s, are 6 electron anionic donors in the ionic model and 5 electon donors in the covalent model and are considered tridentate ligands, and so are referred Figure 2 shows one of the five equivalen t resonance structures for Cp and one of the three equivalent resonance structures for Tp explaining their status as 5 electron donors in the covalent model Like Cp ligands, scorpionate ligands are structured so that they can only bind facially, rather t han meridonally, to a metal center. However, unlike Cp, Tp binds at nitrogen atoms rather than carbon atoms.
3 Tp is a stronger sigma donor and weaker pi accepto r than Cp, and so has a much weaker ligand field. This is evidenced by the fact that Fe(Cp) 2 i s low spin up to temperatures greater than 1000 o C, while Fe(Tp) 2 undergoes a transition from low spin to high spin at approximately 120 o C. 7 Figure 2 : One of seve ral equivalent resonance structures of a Cp donor (left) and a Tp donor (right) to a metal center The steric properties of Tp can be tailored by introducing substituents at the 3 position of the pyr azole rings. Variation of the sub s tituent at the 3 p osition has a large effect on the coordination geometry at a metal center. For example, binding of either unfunctionalized Tp or the 3 methyl, 5 methyl substituted trispyrazolylborate (Tp*) enforces an octahedral g eometry at a metal center, often coordina ting 2:1 to a metal center ( Figure 3 ), while binding of the 3 tert butyl, 5 tert butyl substituted Tp (Tp tB u2 ) usually enforces a t etrahedral geometry, and prevents 2:1 coordination ( Figure 4 ) 2 6 Although Tp tB u complexes tend to be four coordinate with first row transition metals many five coordinate bipyramidal complex es containing small chelating ligands have been structurally characterized; one such example is the copper nitrite complex sho wn in Figure 5 8
4 Figure 3 : Tp 2 Zn, one example of a bis ligand complex generated from Tp ligands containing little steric hindrance 1 Figure 4 : Tp tBu ZnCl, one example of a distorted t etrahedral complex generated from bulky Tp ligands 9 Figure 5 : Tp tBu Cu 2 NO 2 ), an example of a 5 coordinate complex containing a sterically hindered Tp tBu ligand 8
5 It has been found that the steric demand of Tp derivatives increases based on the substituent at the 3 position in the order H < Me < Ph < iPr < tBu. 2 6 The coordination geometry can often be fine tuned through choice of the substituent at the 3 position generally varying between 4 coordinate tetrahedral, 5 coordinate bipyramidal, and 6 coordinate octahedral geometries. One to one r eaction of h ydrotris(3 isopropyl 4 bromopyrazolyl)borate ( Tp iPr ,4Br ) with first row transition metals results in isolation of tetrahedral Tp iPr ,4Br MX complexes (M = Fe, Co, Ni, Cu, Zn), which can further react with Tp or Tp* to fo rm octahedral complexes as shown in Figure 6 10 11 In the tetrahedral complexes, the isopropyl groups rotate so that the methyls are directed toward the metal, while in octahedral complexes, the methyls are directed away from the metal. Reaction of Tp iPr,4Br MX with the slightly bulkier Tp Ph resulted in the Tp iPr,4Br MTp ph complexes shown in Figure 6 (right) in which one pyrazole is held pendant so that the metal instead forms an agostic interaction with the B H bond. The bromo substituent is introduced primarily because it was found that bromo substituted complexes crys tallized more readily. Figure 6 : First row Tp iPr,4Br transition metal complexes exhibit [N 3 ] coordination with the unhindered ligand Tp (left), and [N 2 H] coordination with the more hindered Tp Ph (right) 10 11 (Adapted from ref. 2 )
6 These results are in contrast to the results for Tp tBu which has not had any six coordina te bis ligand complexes characterized for first row transition metals. Because the tert butyl group is not able to rotate so that all of the methyl groups are directed away from the metal, the sterics near the metal center become too congested to accommod ate another scorpionate ligand. In some cases, five and six coordinate complexes containing sterically hindered Tp ligands, in which very unhindered ligands occupy the remaining coordination sites, have been characterized. The six coordinate iron complex shown in Figure 7 has also been structurally characterized, indicating Tp iPr2 allows enough room around a metal center for multiple small ligan ds such as a benzoate molecule and a molecule of acetonitrile solvent 12 Introduction of a substituent at the 5 position helps to protect the boron center from being degraded by a nucleophile, and often simplifies preparation of the ligand. Figure 7 : An example of a sterically hindered Tp iPr2 Fe co mplex exhibiting a 6 coordinate structure in which a molecule of acetonitrile solvent and a molecule of benzoate are coordinated in addition to Tp iPr2 (Taken from ref. 2 ) Scorpionates are useful in that they are versatile ligands with a wide range of
7 applications, and are usually very convenient to use because their synthesis tends to be quite straightforward. Most boron centered homoscorpionates ca n be prepared simply by a neat reaction of sodium or potassium borohydride and a pyraz ole or other similar azole heterocycle such as imidazole or triazole, in a melt. 1 2 Scorpionate complexes have been structurally characterized for almost all metals on the periodic table. Like Cp complexes, scorpionat e complexes have been applied in many areas of chemistry, including polymer ization catalysis, organic couplings, and materials synthesis. T hey have also seen numerous applications in biological inorganic chemistry; i t is only fitting that an i deal application of ligands with such a biologically influenced name is the structural and functional modeling of biological enzyme active sites. Much of this introductory chapter, including the thesis project described herein, is concerned with the applications of scorpionate ligands in synthetic enzyme active site model chemistry. Derivatives of Tp can be used to model the triad of three facial histidine residues found in many enzymes. Complexes containing Tp ligands have been used to model the active sites of zinc enzymes such as carbonic anhydrase and dihydroorotase 13 14 copper enzymes such as hemoc yanin and nitrite reductase 8 15 and iron enzymes such as ri bonucleotide reductase and methane monooxygenase 16 as just a few examples. In addition, models of various cobalt, manganese, nickel, chromium, and molybdenum enzymes have been synthesized 6 One of t he most well studied application s of Tp ligands in the modeling of enzyme active sites is that of carbonic a nhydrase (CA), a ubiquitous zinc enzyme responsible for catalyzing the reversible conversion of car bon dioxide to bicarbonate. 17 Carbonic a nhydrase is of particular interest in part because of its extremely fast turnover rate
8 relative to most enzymes, which is only limited by the rate of diffusion. T he structure of the CA active site is shown in Figure 8 in comparison to the fi r st structural model based on a Tp system in Figure 9 This system was first proposed as a m odel for CA by Alsfasser, Parkin and Vahrenkamp in 1991. 13 Figure 8 : Representation of the active site of the enzyme carbonic anhydrase in its resting state (Taken from ref. 17 ) Figure 9 : Tp tBu,Me ZnOH, the first structural model for the active site of carbonic anhydras e (Taken from ref. 17 )
9 The compl ex Tp tBu,Me ZnOH has been shown to behave as both a structural and functional mi mic for carbonic anhydrase. It was the first structurally characterized mononuclear, tetrahedral zinc hydroxide complex. In addition to having a similar [N 3 O] tetrahedral bind ing scheme with CA, Tp tBu,Me ZnOH also has similar bond lengths and bond angles to CA. The tert butyl substituents at the 3 position of the pyrazolyl rings are necessary to prevent the ligand from coordinating twice to a zinc center. Earlier work with oth er tridentate, facially coordinating [N 3 ] ligands such as 1,4,7 triazacyclononane ( TACN ) instead ga ve complexes such as 6 coordinate 2:1 ligand to metal complexes or 4 and 5 coordinate hydroxyl bridged dim ers. 18 19 The mechanism of CA, shown in Scheme 1 has been studied in significant depth since its structural characterization, and is largely agreed upon. On bi nding of a water molecule to the tetrahedral zinc center, the positive charge at the zinc center lowers the pKa of the water molecule to near biological pH, resulting in deprotonation to form the z inc hydroxide species. The deprotonation is facilitated by a proton shuttle mechanism involving Thr 199 and Glu 106 shown in Figure 8 20 The hydroxide then nucleophilically attacks a molecule of CO 2 resulting in a zinc bound bicarbonate intermediate. A new water molecule then quickly displaces the weakly bound bicarbonate. Although the enzyme has been structurally characterized when bound to both hydroxide and water, the bicarbonate intermediate has not, as of yet, be en characterized. 21 One use of synthetic model studies, therefore, is to obtain useful information on the nature of the zinc bound bicarbonate intermediate.
10 Scheme 1 : Proposed mechanism for conversion of CO 2 to HCO 3 with carbonic anhydras e (Taken from ref. 17 ) Upon exposure of Tp tBu,Me ZnOH to CO 2 Tp tBu,Me ZnOCOOH is formed, as evidenced by IR bands at 1302 and 1675 cm 1 ( Figure 10 ) 22 This complex could not be structurally characterized by X ray diffraction because it crystallized out of solution as the less soluble carbonate bridged dimer Tp tBu,Me ZnOCOOZnTp tBu,Me shown in Figure 11 which was characterized via X ray diffraction. The carboxylate bridged complex was still of significant interest because a carbonate ion had not previously been seen as a monodentate bridging ligand. The complex Tp tBu,Me ZnOCOMe ( Figure 12 ) prepared by treating the zinc hydroxide complex with dimethylpyrocarbonate, was, however, structurally characterized. 23 This complex was found to contain a tetrahedral Zn center with a methyl carbonate ligand bound at one oxygen atom. Tp tBu,Me ZnOCOMe showed IR bands at 1689 and 1297/1280 cm 1 similar to those measured for Tp tBu,Me ZnO COOH, indicating that Tp tBu,Me ZnOCOOH has a similar tetrahedral structure. Both Tp tBu,Me ZnOCOMe and Tp tBu,Me ZnOCOOH are readily hydrolyzed by water. The 1 H NMR signal corresponding to the OH proton disappears on addition of CO 2 and
11 reappears on removal o f CO 2 indicating that the reaction of Tp tBu,Me ZnOH with CO 2 is both rapid and reversible, as is necessary for an effective model of CA. Figure 10 : Solution phase structure of Tp tBu,Me ZnOCOOH prepared from addition of CO 2 to Tp t Bu,Me ZnOH 22 (Taken from ref. 17 ) Figure 11 : Solid state structure of Tp tBu,Me ZnOCOOZnTp tBu,Me isolated from reaction of CO 2 w ith Tp tBu,Me ZnOH (Taken from ref. 17 ) Figure 12 : Solid state structure of Tp tBu,Me ZnOCOMe isolated from reaction of dimethylpyrocarbonate and Tp tBu,Me ZnOH (Taken from ref. 17 ) The use of less sterically hindered Tp ligands yield s Zn complexes that are ineffective as CA models. The slightly less hindered ligand Tp iPr2 forms the dinuclear complex shown in Scheme 2 in which a carbonate bridges two Zn atoms while monodentate to one and
12 bidentate to the other. A tetrahedral zinc bicarbonate species containing only one zinc center is not present in solution, as evidenced by the lack of sim ilar IR bands to those seen for Tp tBu,Me ZnOCOOH In this case, binding of CO 2 is not reversible, and the complex is not hydrolyzed by water. Scheme 2 : In contrast to the reversible reaction of Tp tBu,Me ZnOH with CO 2 Tp iPr2 ZnOH u ndergoes irreversible dimer formatio n (Taken from ref. 17 ) In order to study the conversion of the zinc aqua complex to the zinc hydroxide complex, it was necessary to prepare Tp tBu,Me ZnOH 2 + Characteri zation of Tp tBu,Me ZnOH 2 + p roved much more challenging than characterization of Tp tBu,Me ZnOH, as protonation of a bound
13 hydroxide ion immediately resulted in irreversible displacement of the water molecule by an anion. 24 It was, however, found that reaction with (C 6 F 5 ) 3 BOH 2 resulted in the formation of a complex that could be structurally characterized, in which Tp tBu,Me ZnOH 2 + is hydrogen bonded to the counterion (C 6 F 5 ) 3 BOH ( Scheme 3 ). 25 The hydrogen bonding interaction with (C 6 F 5 ) 3 BOH is analogous to the hydrogen bonding interaction with Thr 199 in the actual enzyme, which is necessary for conversion between th e aqua and hydroxide species. NMR studies show that interconversion between Tp tBu,Me ZnOH and Tp tBu,Me ZnOH 2 + is fast on the NMR timescale. This facile proton exchange is again necessary to effectively model the mechanism of CA. It has been found that bin ding of water in this system lowers the pKa of the bound water from 15.7 to approximately 6.5, which is close to physiological pH, which is also necessary for CA to be able to effectively carry out its function. As would be expected, the protonated comple x is not a good enough nucleophile to react with CO 2 providing evidence that CA must also be deprotonated in order to be an effective catalyst. Scheme 3 : Tp tBu,Me ZnOH 2 + a model for protonated CA, was first prepared by reaction with (C 6 F 5 ) 3 B OH 2 in Et 3 N (Taken from ref. 17)
14 Isotope exchange studies have been carried out to determine whether the complex Tp tBu,Me ZnOH can model the final step in the mechanism of CA, which is displacement of the bound bicarbonate by a water molecule. Using 17 O NMR studies, it was found that 17 O is able to rapidly exchange between H 2 17 O and CO 2 affirming that Tp tBu,Me ZnOH does perform the function of CA. 22 Thus, Tp tBu,Me ZnOH is the first complex a nd still one of few complexes to model both the structure and function of CA. This system presents an excellent example of how the sterics of a scorpionate ligand can be precisely tuned to give complexes with desired structures and reactivities. Boron cen tered scorpionate complexes have also been synthesized from heterocycl ic precursors other than functionalized pyrazoles. The most studied of these has been Hydrot ris( mercaptoimidazolyl ) borate (Tm) first reported by Regl inski and coworkers in 1996 ( Figure 13 ). 26 Like Tp, Tm generally coordinates facially to metal centers, but it coordinates at softer sulfur atoms relative to the nitrogen atoms in Tp. It has hence been consi As would be expected, Tm shows significantly less ability than Tp to coordinate to hard metals like Mg and Ca, while binding quite strongly to soft er metals like Zn and Hg. 27 Tm ligands have been found to be considerably more electron donating than Tp ligands based on comparison of IR stretching frequencies of analogo us metal carbonyl complexes 28 As with Tp liga nds, the steric demand of Tm R ligands, and thus the mode of binding and available coordination sites, can be tuned through choice of the substituent at the 1 position of the imidazole ring represented by the superscript in Tm R As for Tp, complexes of Tm Me tend to coordinate 2:1 to first row transition metals, resulting in distorted octahedral geometry, while the bulkier Tm tBu favors tetrahedral metal complexes. Coordination of Tm ligands results in an eight
15 membered chelate ring, rather than the six mem bered ring found in Tp complexes. This affords somewhat more flexibility, not strictly limiting Tm complexes to boat like chelate ring structures. 29 Figure 13 : An [S 3 ] monoanio nic h ydrotris mercaptoimidazolylborate (Tm R ) ligand Bulky Tm ligands such as Tm tBu have been used to create tetrahedral [S 3 ] environments around metals for applications in enzyme modeling. Zinc complexes have been shown to behave both as structural and f unctional ana logs for organomercurial lyase 30 and the ADA DNA repair protein 31 both of which contain tetrahedral zinc centers support ed by three cystei ne residues. Heteroscorpionates have also been used in active site models The [S 2 N] zinc compound shown in Figure 14 has been applied in the modeling of liver alcohol dehydrogenase, an enzyme in which a tetrahedral zinc center is bound by two cysteine resid ues and one histidine residue. 32
16 Figure 14 : A heteroscorpionate zinc complex containing two mercaptoimidazol donors and one pyrazol e donor used to model liver alcohol dehydrogenase (LADH ) (Taken from ref. 29 ) Recently, the Parkin group has developed new imidazole scorpionate ligands in which the sulf ur donor is replaced by a selenium atom 33 or an oxygen atom 34 shown in Figure 15 Like Tm R Hydrotris(selenoimidazolyl)borates ( Tse R ) Hydrotris(oxoimidazolyl)borates ( To R ) and heteroscorpionates containing combinations of pyrazole, mercapto seleno and/or oxoimid azole may also have potential applications in bioinorganic chemistry modeling active sites of enzymes containing residues such as selenocysteine and serine. Figure 15 : Hydrotris( selenoimidazolyl ) borate (Ts e R ) and Hydrotris( oxoimidazolyl ) borate (To R ), recently published [Se 3 ] and [O 3 ] homoscorpionate ligands analogous to the more commonly used Tm R
17 1.2 Carbon Centered Scorpionate Ligands In 1970, Trofimenko introduced tris pyrazolylmethane (tpzm), a neutral carbon cente red analog of Tp ( Figure 16 ) 35 Figure 16 : A tris(pyrazolyl)methane (Tp ) ligand, a neutral analog to the monoanionic Tp ligand Tpzm is a tridentate, 6 electron neutral donor isolobal to neutral 6 arene ligands such as benzene. As with Tp, the steric requirements of tpzm ligands can be tuned by adjusting the substituents at the 3 po sition of the pyrazole rings. T pzm complexes generally show very similar preferences in coordination geometry (i .e. octahedral, bipyramidal, or tetrahedral) to analogous Tp complexes. 36 They also provide a similar ligand field and are only slightly less donating than Tp ligands. 37 Tpzm ligands initially saw very little attention largely due to their difficulty of synthesis relative to other [N 3 ] ligands such as Tp and TACN In 198 2, Elguero and coworkers reported an efficient method to produce bis tris and te tra kis azolylmethanes in high yield from substituted and unsubstituted azoles, including some imidazoles and triazo les in addition to pyrazoles. 38 39 Th is synthesis made tpzm ligands much more accessible, so that their use as neutral, tridentate [N 3 ] ligands became much more frequent among inorganic chemists, although n ot nearly
18 to the extent of Tp. Elguero Scheme 4 The polyazole is prepared by heating n equivalents of azole in excess CH 4 n Cl n in the presence of base under phase transfer catalysis conditions; where n is two for bis azolylmethanes, three for tris azolylmethanes, and four for tetrakis azolylmethanes. The reaction can be carried out either under liquid liquid phase transfer conditions, in which base is dissolved in a separate aqueous layer, or under solid liquid phase transfer conditions, in which there is only an organic l ayer, while solid base is stirred in the solution. In either case, a phase transfer catalyst facilitates the transport of hydroxide ions between phases in order to deprotonate the acidic nitrogen on the azole ring to e ffect nucleophilic attack of the chlo ride solvent. Scheme 4 : General synthesis for poly azolylmethanes proposed by Elguero and coworker s (Adapted from ref. 39 ) Tpzm ligands offer several potential advantages over analogous Tp ligands, particular ly in areas related to enzyme modeling. First, the C N and C H bonds at the central carbon are generally much less reactive than the B N and B H bonds found in Tp. Although the B N bonds are usually stable, they have b een shown to hydrolyze particularly when there is not significant steric protection at the 5 position of the pyrazole rings, while the C N bonds are not readily hydrolyzed. 36 Additionally, the B H bond ofte n coordinates to metal centers as a sigma donor in place of a pyrazolyl nitrogen, which can occasionally even lead to hydride transfer from boron to the metal. 40 While this behavior is often
1 9 interesting, it is not desirable when the ligand is being used as a tridentate [His 3 ] mimic in synthetic enzyme models. Second, t pzm is neutral rather than monoanionic, coordinating to a metal with three dative bonds, which is analogous to the manner in which three f acial histidine residues bind a metal. The absence of a formal charge on the ligand does, however, generally make tpzm ligands both more labile and more weakly donating than Tp ligands, so that they are more readily displaced by other ligands 36 Another advantage of tpzm ligands is that the central carbon can be functionalized in a number of different ways by replacing the C H. A much wider range of stable functionalizations can be achieved with a centra l carbon than with a central boron. Most functionalizations are achieved by deprotonating the C H bond with a strong base such as butyllithium or potassium tert butoxide, followed by introduction of an electrophile. For example, an alcohol functional gro up can be added by deprotonation with potassium tert butoxide followed by reaction with formaldehyde, as shown in Scheme 5 41 Scheme 5 : Tpzm ligands can be readily functionalized by deprotonation of the central carbon with a strong base followed by reaction with an electrophilic substrate; in this case, formaldehyde (Taken from ref. 36 ) One potentially useful application of functionalizing the central carbon has been the synthesis of coordination polymers, prepared by covalently linking multiple tpzm ligands functionalized with an alcohol group before introduction of a metal, such as c admium, as
20 shown in Scheme 6 These coordination polymers have intricate supramolecular structure s that differ significantly based on whether tpzm ligands are joined via ortho, meta, or para linkages, due to facto rs such as hydrogen bonding and pi stacking interactions. 41 Scheme 6 : Several cadmium coordination polymers can be prepared using tpzm ligands linked through a phenyl grou p (A dapted from ref. 41 ) Despite being structurally analogous to enzyme active sites incorporating three facial histidine residues coordinated to a metal center, both tpzm and Tp complexes have a disadvantag e in enzyme modeling in that they are generally insoluble in aqueous solution. The hydrophobic nature of the ligand s makes it difficult to study the reaction chemistry of complexes containing these ligands in aqueous media. Even charged tpzm complexes ar e generally insoluble in aqueous solution. In contrast, most enzymatic reactions take place in the presence of significant amounts of water. Additionally, insolubility of Tp and tpzm complexes in water makes it impossible to determine pKa of basic or aci dic complexes such as the previously discussed Tp tB u ZnOH used as a model for carbonic anhydrase ( Figure 9 ) by standard means Even if a water soluble Tp complex were prepared, the B N bonds could easily be hydro lyzed by water, resulting in
21 decomposition of the ligand. In an attempt to design a water soluble [N 3 ] ligand for enzyme active site modeling, Klaui and co workers synthesized the ligand t ris 3 tert butyl pyrazol ylmethanesulfonate (tpms tB u ) via Scheme 7 42 Scheme 7 : The water soluble tris pyrazolylmethanesulfonate ligand can be prepared from treatment with sulfur trioxide after deprotonati on of the central carbon (Taken from ref. 42 ) In theory, the highly hydrophilic, yet generally weakly binding sulfonate group would be held pendant while the three pyrazole rings bound to a metal cente r so that the sulfonate could interact with aqueous solvent It was found that the lithium salt of the sulfonate was both water soluble and stable over the pH range of 0 to 13. However, studies showed that tpms tB u often tends to bind in a facial [N 2 O] s cheme, binding at a sulfonate oxygen rather than a pyrazole nitrogen, for biologically r elevant metals such as zinc, copper and cobalt 43 The ligand does, however tend to bind [N 3 ] to some metals, such as nickel and vanadium. 1 H NMR studies in CDCl 3 showed an equilibrium b etween [N 3 ] and [N 2 O] tpms tB u ZnCl ( Scheme 8 ), with [N 2 O] binding favored at higher temperature and [N 3 ] binding favored at low er temperature.
22 Scheme 8 : An equilibrium exists between [N 3 ] and [N 2 O] tpms tBu ZnX (X = Cl, Br Et ) in solutio n (Taken from ref. 43 ) Based on this information as well as calculations Klau i argued that the [N 3 ] complex is favored enthalpically due to stronger interactions between the pyrazole nitrogens and the zinc center, while the [N 2 O] complex is favored entropically due to a smaller dipole moment allowing more disorder in the solvent. Only [N 2 O] tpms tB u ZnBr was able to be characterized in the solid state via X ray diffraction, and its structure is shown in Figure 17 Figure 17 : ORTEP diagram of tpms tBu ZnBr in which t he ligand is bound [N 2 O] rather than [N 3 ]. (Taken from ref. 43 )
23 Interestingly, reaction of [N 2 O] tpms tBu ZnEt with 0.5 equivalents of acetic acid gives the acetato and oxo bridged dinuclear complex shown in Figure 18 which is one of very few structural models for the dinuclear zinc enzymes phospholypase C and phosphotriesterase. Figure 18 : ORTEP diagram of acetato bridged dimer prepared from tpms tBu ZnEt and 0.5 equivalents acetic acid (Taken from ref. 43 ) 1.3 Bis Pyrazolyla cetate s and Related [N 2 O] Ligands Facial [ N 2 O] binding of two histidine residues and either a glutamate or asp artate residue to a metal center is a very common structural theme in enzyme active sites, most notably in various zinc p roteases and iron oxidas es 44 Z inc proteases selectively hydrolyze specific peptid e bonds, while iron oxidases oxidize specific substrates. Significant work has gone into preparing heteroscorpionate ligands that successfully model this binding scheme. Initial attempts at [N 2 O] ligands resulted in complexes that did not have ligands bo und in the appropriate facial orientation. For example, it was found that bis[(3,5
24 diisopropylpyrazolyl)ethyl]ether binds to zinc meridonally rather than facially, as shown in Figure 19 45 Figure 19 : B is[(3,5 diisopropylpyrazolyl)ethyl]ether is an [N 2 O] ligand that binds meridonally, rather than facially, to zinc (Taken from ref. 45 ) A tetrahedral [N 2 O] z inserting formaldehyde into the B H Zn bond in the tetrahedral complex Bp tBu,iP r ZnCl ( Scheme 9 ). 46 This method was expanded upon by inserting carbon dioxide rather than formaldehyde to create a closer approximation to a carboxylate donor. 47 This complex is a good structural model for zinc proteases such as thermolysin (TLN), carboxypeptidase b (CPB), and angiotensin converting enzyme (ACE). In 1999, Carrano and coworkers first prepared the carbon centered [N 2 O] ligand (3 tert butyl 2 hydroxy 5 methylphenyl) bis(3,5 dimethylpyrazo l 1 yl)methane. 48 This ligand was also shown to support mononuclear tetrahedral zinc centers ( Figure 20 ).
25 Scheme 9 : The [N 2 O] facial b inding scheme common in enzyme active sites was first accessed by inserting carbon dioxide or ketones /aldehydes into the B H Zn bond in Bp complexe s. (Taken from ref. 17 ) Figure 20 : (3 T ert butyl 2 hydroxy 5 methylphenyl) bis(3,5 dimethylpyrazol 1 yl)methane binds as an [N 2 O] facial ligand to produce tetrahedral zinc complexe s. (Taken from ref. 17 ) Although the complexes prep structural models for [N 2 O] supported zinc peptidases, there are several ways in which
26 they could be improved. tetrahedral transition met al species in which one coordination site is occupied by a B H sigma bond. The B H bond must be activated enough for carbon dioxide to insert. Hence, only a limited number of [N 2 O] supported transition metal species could be accessed by this route. Seco nd, neither ligand bind s at a carboxylate group, although enzyme containing a [His,His,Tyr] binding scheme, which are not common in enzyme active sites. Carboxylate oxygens exhibit significantly diffe rent coordination properties fro m, for example, phenoxide ligands and therefore exert different structural and electron ic effects on a metal center. A carboxylate moiety also has the potential to bind 1 2 ). Many enzymes, including carboxypeptidase A, have active sites in which a zinc center is bound by both carboxylate oxygens. 44 In 1 and 2 in order to carry out their functions. dimethylpyrazol 1 yl)acetate (bdmpza) the fi rst ligand shown to bind facially to metal centers at two nitrogens and a carboxylate oxygen via the same route as that shown for the tert butyl derivative in Scheme 10 49 This ligand was prepared in the same manner as the trispyrazolylmethanesulfonate ligands described earlier 50 B is(3,5 dimethylpyrazol 1 yl)methane was prepared from methylene chloride, 3,5 di methylpyrazole, and base, then deprotonated with butyllithium and treated with carbon dioxide. Not only is this synthesis relatively simple and efficient, but it gives an [N 2 O] ligand that is one of, if not
27 the closest synthetic approximation to date for the [His,His,Glu/Asp] binding scheme in enzymes. 44 Scheme 10 : Two step synthesis of Hbdtbpza in which bdtbpzm is prepared, and then functionalized with CO 2 (Taken from ref. 44 ) In 2001, Burzlaff and coworkers introduced the first zinc complexes of this ligand. He devised an altered synthesis of bdmpza that required only one step, in which either dibromoacetic acid or dichloroa cetic acid rather than dichlorome thane reacts with 3,5 dimethylpyrazole under similar conditions ( Scheme 11 ) 51 Scheme 11 : Alternative synthesis of Hbdmpza (R = CH 3 ) and Hbpza (R = H). The one step (Taken from ref. 44 ) As expected based on earlier r esults with the slight ly more hindered ligands Tp* and tris 3,5 dimethylpyrazolylmethane ( tdmpzm ) only the mononuclear bis ligand zinc complex ( Figure 21 ) could be isolated, even at 1:1 metal to ligand ratio or hi gher.
28 Figure 21 : ORTEP diagram of Zn(bdmpza) 2 in which two equivalents of ligand bind to an octahedral zinc center (Taken from ref. 44 ) In order to prepare tetrahedral com plexes in which only a single ligand is coordinated, Burzlaff synthesized bis(3,5 di tert butylpyrazol 1 yl)acetic acid ( H bdtbpza) via the route initially used by Otero for bdmpza applied to sterically hindered pyrazoles. It was found that bdtbpza coordinated only once to first row transition metals, and the tetrahedral complex Zn (bdtbpza) Cl was readily prepared from reaction of the deprotonated ligand with zinc chloride as shown in Scheme 12 Scheme 12 : Reaction of bdtbpza with zinc chloride results in a tetrahedral zinc complex containing one equivalent of ligan d. (Taken from ref. 44 )
29 This complex presented the best structural model to date for tetrahedral zinc proteases exhibiting [N 2 O] coordination in which only one carboxylate oxygen is coordinating, such as TLN, CPB, and ACE. The O Zn N and N Zn N bond angles are similar to those O bond is significantly shortened from 2.065 (4) to 1.990 (2) a distance which is more characteristic for carboxylate donors. In addition to Zn (bdtbpza) Cl, Zn (bdtbpza) Me was prepared from reaction of Hbdt bpza with dimethylzinc This compound offers an advantage over the chloride complex in that the methyl group is much more easily exchanged. I n many circumstances, treatment of tetrahedral methyl zinc species supported by ligands such as Tp tBu 2 tdtbpzm, and bdtbpza with protic compounds readily results in substitution of the methyl ligand accompan ied by liberation of methane It was found that the chiral scorpionate ligand (3 ,5 di tert butylpyrazol 1 dimethylpyrazol 1 yl)acetic acid (HBpa tBu 2,M e2 ) also coordinated only once to zinc centers, resulting in zinc complexes that exhibit chirality as in enzyme active sites. 3 An ORTEP diagram of Bpa tBu 2,Me2 ZnMe is shown in Figure 22 Figure 22 : ORTEP diagram of Bpa tBu2,Me2 ZnMe, a chiral, tetrahedral bis(pyrazolyl)acetate complex (Taken from ref. 44 )
30 The complex cation, Bpa t Bu2,M e2 ZnOH 2 + up and is an excellent structural model for the resting state of [N 2 O] supported zinc proteases, exhibiting both the correct facial binding scheme and a chiral center. 52 Despite its existence in solution, this compound could only be isolated as a carboxylate bridged polymer in the solid state, as shown in Figure 23 Hydrogen bonding interactions between the aqua ligand a nd the carboxylate likely promote the formation of polymers. Figure 23 : ORTEP d iagram of Bpa tBu2,Me2 ZnOH 2 + whic h is a carboxylate bridged polymer in the solid stat e. (Taken from ref. 44 ) Inhibition of proteases such as ACE using inhibitors containing zinc binding groups (ZBGs) is important in many aspects of medicine and biology, including treatment of hypertension. 44 In an effor t to study the coordination prop erties of several ZBGs with applications in zinc protease inhibition, Burzlaff have studied several substituted zinc complexes originating from both the chloride and methyl species of either the Bpa t Bu2,Me2 or bdtbpza ligand The first of such species prepared was the
31 zinc thiolate compound (bdtbpza)ZnSCH 2 Ph prepared via the reaction shown in Scheme 13 51 Isolation of the product proved difficult, however, as compared to preparations that originated from the zinc methyl species. Future studies generally utilized Bpa tBu2,Me2 ZnMe or (bdtbpza)ZnMe as precursors to the target complexes. Scheme 14 shows the syntheses of some 5 coordinate zinc complexes containing the ZBGs thiolato, hydroximato, or carboxylato from zinc alkyl precursors as models for inhibited zinc protease enzymes. Scheme 13 : Preparation of (bdtbpza)ZnSCH 2 Ph by reaction of (bdtbpza)ZnCl with HSCH 2 Ph under basic condition s (Taken from ref. 44 )
32 Scheme 14 : 5 C oordinate zinc complexes that model inhi bitors bound to proteases have been prepared from (bdtbpza)ZnM e. (Taken from ref. 44 ) As mentioned earlier, the [N 2 O] facial binding scheme is also quite common in iron oxidase enzymes in addition to zi nc proteases. While these iron oxidases are supported with the same binding scheme as the zinc proteases discussed earlier, these enzymes generally contain an octahedral rather than tetrahedral iron center. 44 The main function of such enzymes is the catalytic oxidation of some substrate using molecular oxygen. In the resting state of most iron oxidases, the iron center is in the Fe(II), or ferrous, oxidation state. Throughout the course of the reaction the iron atom changes between multi ple oxidation states, including Fe(II), Fe(III) (ferric) and Fe(IV) as oxygen binds and reacts with the substrate. Even in the Fe(II) oxidation state, the active sites tend to be high spin because the ligand field crea ted by the [N 2 O] binding is relatively weak. As
33 with zinc protease enzymes, it has been found through site directed mutagenesis studies that the [N 2 O] triad is essential for enzyme activity; substrate conversion is reduced significantly if, for example, t he carboxylate donor is replaced with a third histidine nitrogen donor. One major class of iron oxidase enzymes exhibiting this donor array is the 2 oxoglutarate dependent Fe(II) oxidases. 53 The general reaction catalyzed by thes e oxidases is shown in Scheme 15 54 Some examples of such enzymes are taurine dioxygenase and prolyl 4 hydroxylase, responsible for sel ective hydroxylation of taurine and proline respectively. These enzymes are capable of specifically oxidizing a C H bond of a substrate to a C OH bond using a molecule of oxygen and the cosubstrate 2 oxoglutarate. In the process, a molecule of 2 oxogluta rate is converted to succinate and carbon dioxide. A proposed mechanism for this transformation is shown in Scheme 16 Scheme 15 : Many iron oxidases catalyze hydroxylation of a substrate coupled to conversion of 2 oxoglutarate to succinate and CO 2 (Taken from ref. 44 )
34 Scheme 16 : General mechanism for 2 oxoglutarate dependent iron oxygenase enzymes (Taken from ref. 44 ) In the resting state, three water molecules are bound to the iron center in addition to the two histidine nitrogens and one carboxylate oxygen resulting in a slightly distorted octahedral g eometry. Two of the labile aqua ligands are displaced by a molecule of 2 oxoglutarate A molecule of substrate diffuses into a binding pocket near the iron center, but does not bind directly to the iron. Dissociation of the third water molecule results in a 5 coordinate structure. A molecule of oxygen likely inserts into this species to form a superoxide structure in which the Fe(II) center is oxidized to Fe(IV). The Fe(IV) oxo compound then forms on decarboxylation of the bound 2 oxoglutarate to form a molecule 2 succinate molecule. Formation of CO 2 is irreversible
35 and is likely the driving force for the reaction. The iron oxo intermediate then abstracts a hydrogen atom from the substrate and oxidizes it to an alcohol, resulting in reduction of the iron center back to Fe(II). The weakly bound succinate is then displaced from the active site by water molecules to complete the catalytic cycle. Creating structural and functional models for iron oxidases may be useful both in understanding the reaction mechanisms of such enzymes and in creating small molecule catalysts for conversion of C H bonds to C OH bonds. Thus, significant work has gone into using [N 2 O] ligands, particularly bispyrazolylacetates, to model iron oxidase s. As with the analogous zinc complex, reaction of FeCl 2 with bdmpza results on ly in isolation of the bis ligand complex (bdmpza) 2 Fe. 51 This is a n air stable, high spin 18 electron species Even at te mperatures as low as 5 K, SQUID magnetometry data show that the complex is high spin, indicating the ligand has a much weaker ligand field than tdmpzm or Tp*. Exchange of a single bdmpza ligand was found to not be feasible, so bdtbpza was used in an attem pt to isolate a mono ligand species instead. Under similar reaction conditions, this resulted in isolation of a five coordinate dimer in the solid state, containing one molecule of ligand per iron atom, in which t w o iron centers are bridged by a single ca rboxylate oxygen ( Scheme 17 )
36 Scheme 17 : Reaction of bdtbpza with ferrous chloride results in a carboxylate bridged dimer in the solid state (Taken from ref. 44 ) The compound [Fe(bdtbpza)Cl] 2 is very air sensitive; like iron oxygenase enzymes, it is readily oxidized by molecular oxygen. The dimer can be broken up into an isolable 5 coordinate monomer by introduction of a bidentate lig and such as benzoylformate. 55 Reaction with thallium benzoylformate gives the compound shown in Figure 24 This complex is a very good structural model for iron oxygenase enzymes bound to a 2 oxoglutarate cosubstrate, as in Scheme 16 The benzoylformate l igand models the bound 2 oxoglut arate cosubstrate. It has been shown to be readily oxidized by molecular oxygen. The compound shows a strong metal ligand charge transfer absorption band at 530 nm, as is also seen in the resting state of iron oxidases such as taurine dioxygenase. The band immediately disappears when the compound is oxidized by being exposed to air. Des pite being a very good structural model, studies do not show that the compound [Fe(bdtbpza)(O 2 CCOPh)] acts as a functional model. Exposure to oxygen did not result in formation of benzoic acid, as would be expected if the iron complex was a functional mim ic for 2 oxoglutarate dependent iron oxygenases ( Scheme 16 ).
37 Figure 24 : The 5 coordinate iron compound isolated from reaction of [Fe(bdtbpza)Cl] 2 with benzoylformate is a good structura l model for iron oxygenase enzymes bound to a succinate molecul e. (Taken from ref. 44 ) Interestingly, reaction of bdtbpza with an Fe(II) source containing the weakly coordinating tetrafluoroborate count erion resulted in isolation of the bis ligand complex Fe(bdtbpza) 2 which is similarly unreactive and air stable to Fe(bdmpza) 2 Hence, in contrast to tris 3,5 di tert butylpyrazolylmethane ( t d tbpzm ) and Tp tBu2 both mono and bis ligand iron complexes ca n be isolated for bdtbpza This is a consequence of bdtbpza having relatively low steric hindrance due to only containing two, rather than three, sterically encumbering tert butyl groups positioned near the iron center, allowing enough room for the bis li gand complex to form under some conditions. 1.4 Bdippza: A new ligand for modeling RuBisCo The goal of the project described herein is to use a bpza ligand to structurally and functionally model the active site of ribulose 1,5 bisphosphate carboxylase ( RuBisCo ) RuBisCo is responsible for catalyzing carbon dioxide fixation, or conversion of carbon dioxide to an organic molecule. In plants, this is the first step in converting carbon
38 dioxide to good sources of energy, such as carbohydrates. RuBisCo i s noteworthy for having a very slow turnover rate relative to most enzyme s yet it is also the only system known to catalytically carry out the transformation it facilitates. 56 Because of its low activit y, a very large amount of this enzyme is produced by plants; as such, it is widely cited as the most ubiquitous enzyme on earth. Because of the possible utility of such a system in applications such as sequestration of atmospheric carbon dioxide, there is an interest in designing systems that either mimic or improve upon the catalytic reaction or that provide insight into the mechanism. RuBisCo contains a magnesium ion supported by three facial enzyme bound oxygen atoms at the active site. 56 It is additionally coordinated by three water molecules in its resting state resulting in an octahedral geometry. A proposed mechanism for CO 2 fixation is shown in Scheme 18 In this mechanism, D ribulose 1,5 bisphosphate, a carbohydrate substrate binds to the magnesium ion, displacing two coordinated water molecules (A). The enzyme bound carbamate ligand acts as a general base to effect en edi oliz ation of the carbohydrate ( B). The carbamate effects multiple proton transfers with the aid of a lysine residue positioned close to the active site (C E). This results in a species capable of nucleophilically attacking CO 2 upon addition of water resulting in displacement of a meta l bound aqua liga nd by CO 2 to form a 6 membered chelate (F) Another carbamate induced proton transfer results in cleavage of a C C bond to form the products upper 3 phospho D glycerate and lower 3 phospho D glycerate, which are readily displaced by wate r molecules (G I).
39 Scheme 18 : Proposed r eaction mechanism of RuBisCo. M represents an Mg 2+ center (Taken from ref. 56 ) Previous work in the Sherman lab has focused on the s ynthesis of small molecule analogs to the active site of RuBisCo. Specifically, a mononuclear magnesium complex cation containing the ligand 1,4,7 triazacyclononane N acetate has been synthesized. 57 A c rystal structure of this complex is shown in Figure 25
40 Figure 25 : TACN mono acetate magnesium complex prepared by the Sherman group in initial attempts to model RuBisC o. (Taken from re f. 57 ) The complex is 6 coordinate, with the [N 3 O] TACN monoacetate ligand taking up four of the coordination sites and aqua ligands filling the remaining two. In this structure, the carboxylate arm mod els the binding of the carbamate binding group in the enzyme. T he complex was shown to facilitate slow exchange of deuterium from MeOD solvent for hydrogen in the substrate acetol, a molecule with a structure analogous to D ribulose 1,5 bisphosphate ( Figure 26 ), as evidenced by proton NMR 58 Deuterium exchange likely occurred by going through an enediolate intermediate, as in the initial step of the mechanism of RuB isCo after binding acetol This shows that the magnesium complex is likely able to mimic both the acetol binding and enediolization steps, but does so at a very slow rate, over the span of several days. It was found that other magnesium compounds could c atalyze the exchange much more rapidly in the presence of added
41 base 59 Efforts to crystallize and structurally characterize a complex in which the aqua ligands were displaced by acetol were unsuc cessful. Figure 26 : Acetol can be used as a low molecular weight analog of D ribulose 1,5 bisphosphate, the substrate for RuBisCo, in synthetic model studies. The atoms labeled with the same numbers are analogous to one a nother. This project attempts to model RuBisCo using the ligand bis 3,5 diisopropylpyrazol 1 ylacetate (bdippza), shown in protonated form in Figure 27 While the dimethyl, di tert butyl, and other derivatives of this ligand have been extensively studied by Otero, Burzlaff, Carrano, and others this particular ligand has not yet been reported in the literature. Additionally, a magnesium complex containing a bis pyrazolylacetate ligand has yet to be reported. Alt hough it is less likely to enforce octahedral coordination than TACN mono acetate a tridentate [N 2 O] ligand is a closer approximation to the tridentate [O 3 ] binding scheme at the active site of RuBisCo A tridentate ligand may allow more room in the coord ination sphere for a small bidentate ligand such as acetol to coordinate. Additionally, three additional available coordination site s (assuming octahedral coordination) allows the potential for modeling later steps in the mechanism of RuBisCo, in which ca rbon dioxide binds at the magnesium center. In modeling RuBisCo with
42 either a [N 3 O] or [N 2 O] ligand, coordination sites that are occupied by aspartate and glutamate carboxylate oxygens in the enzyme are instead occupied by nitrogens. A goal of later atte mpts at modeling RuBisCo may, therefore, be to synthesize a tridentate ligand that binds in an [O 3 ] scheme rather than a [N 2 O] scheme. The first goal of the project is to prepare and characterize zinc complexes containing this ligand. While the active sit e of RuBisCo contains a magnesium center rather than a zinc center, zinc complexes are often easier to prepare and crystallize. Magnesium and zinc have similar ionic radii and therefore similar preferences for tetrahedral versus octahedral geometry, since neither experiences ligand field effects. However, Mg 2+ is much harder than Zn 2+ giving it more of a preference to be bound at hard oxygen donors rather than softer nitrogen donors. Figure 27 : The proposed ligand, bdippza, i s intermediate in sterics between bdmpza and bdtbpza As discussed earlier, preparations for tetrahedral, mono ligand zinc complexes containing bulky [N 2 O] ligands like bdtbpza and for octahedral bis ligand complexes containing unhindered ligands such as bd mpza are already known. Therefore, one or both of such
43 complexes should be readily prepared from bdippza and Zn(II) sources. It is expected that because isopropyl groups offer sterics intermediate between methyl and tert butyl, a zinc complex will adopt one of the coordination modes adopted by these species. Ideally, either a tetrahedral or octahedral complex could be obtained depending on reaction conditions, even though this is not the case for either bdmpza or bdtbpza. It is also possible that a mixt ure of species could result. Due to their similar sizes, both Mg 2+ and Zn 2+ complexes are likely to coordinate the same number of bdippza ligands It is, however, possible that bdippza will not coordinate to magnesium with a standard tridentate [N 2 O] sch eme and will instead coordinate at the carboxylate group at one or both oxygens, or in a manner such that the carboxylate bridges two magnesium centers, and at neither or only one of the nitrogens due to its much greater affinity for oxygen over nitrogen. Since a bis ligand complex would likely be very unreactive, the tetrahedral, mono substituted zinc compound ( Figure 28 ), in which X is a halide, alkyl, or solvent molecule, is the desired product. The fourth liga nd can be readily displaced, as in Scheme 13 and Scheme 14 to give 4, 5, or 6 coordinate compounds containing various small monodentate or chelating ligands. Using a ligan d containing isopropyl groups rather than tert butyl groups offers the advantage that higher coordination numbers may be more accessible, but offers the risk that two molecules of the ligand will coordinate irreversibly, resulting in a bis ligand complex, as is the case for bdmpza.
44 Figure 28 : A goal of this project is to prepare tetrahedral zinc complexes of the structure shown A second goal of this project is to prepare and characterize tetrahedral magnesium complexes with the s tructure Mg[bdippza]X. Although magnesium complexes tend to be more difficult to isolate and crystallize than corresponding zinc complexes, a recent study by Mountford and coworkers do give evidence that the ligand will bind to magnesium with a tridentate [N 2 O] binding scheme, rather tha n a scheme in which one or neither nitrogen are bound. 60 In this study, Mountford prepared and characterized several sodium, magnesium and zinc complexes containing the h eteroscorpionate ligand 2,4 di tert butyl 6 (bis(3,5 dimethylpyrazolyl)methyl)phenol Several reactions that the ligand undergoes with sodium and magnesium reagents are shown in Scheme 19 Although there have bee n many [N 3 ] scorpionate magnesium complexes presented in the literature, this study presented the first examples of magnesium complex es containing an [N 2 O] heteroscorpionate ligand in the structurally characterized complexes 3 4, 6 and 7 While the liga nd coordinated in an [N 2 O] tridentate manner in all characterized magnesium
45 complexes, it was found to coordinate at only one nitrogen to sodium, and to bridge two sodium centers at the oxygen to form a dimer in the solid state. This is because the harder sodium center has greater affinity for oxygen and less affinity for nitrogen than magnesium. Scheme 19 : 2,4 D i tert butyl 6 (bis(3,5 dimethylpyrazolyl)methyl)phenol reacts with magnesium and sodium sources to form a variety of tetrahedral and octahedral compound s. (Taken from ref. 60 ) In the syntheses of 3 5 6 and 7 a small to moderate amount of bis ligand magnesium complex 4 was also produced. Although Mountford and co workers were not able to completely separate the two species, they were able to obtain single crystals of each, and have reported X ray crystal structures of 3 4 6 and 7 It was also found that the structurally characterized magnesium complexes had an alogous structures to their
46 corresponding zinc complexes, albeit with shorter M O bonds and longer M C and M N bonds, which would again be expected based on the HSAB principle. Because the pyrazolyl groups are somewhat more hindered in bdippza, but the ca rboxylate group is Both are [N 2 O] ligands with similar electronic properties. The main differences be tween the ligands are that oxygen, and that bdippza contains all 6 membered chelates when coordinated to a metal, membered chelates and one 6 membered chelate. As such, there is a good probability that bdippza will coordinate as a tridentate [N 2 O] ligand to m agnesium in a mixture of the mono ligand tetrahedral complex and bis ligand octahedral complex that is dependent on conditions. In fact, because of the favorable formation of all 6 membered chelate rings, a structure in which the ligand is coordinated [N 2 O] may be even more stable for bdippza. As a side note, Mountford reports in this paper that the magnesium complexes are activ e, although poor, ring opening polymerization (ROP) polymerization catalysts. It may therefore be interesting to investigate similar tetrahedral bdippza magnesium complexes for catalytic behavior toward monomers capable of being polymerized through a ROP mechanism. A third goal of this project is to pr epare structurally characterize, and investigate bdippza zinc and magnesium complexes in which an acetol molecule is bound, either replacing the fourth ligand to form a 5 coordinate complex or leaving it b ound to form a 6 coordinate complex Such a complex particularly one containing magnesium, would be an excellent structural model for the active site of RuBisCo. By studying the exchange
47 rate of the protons o n carbon 2 of acetol ( Scheme 20 ), it can be determined whether the model complexes catalyze en edi olization of acetol as in Scheme 18 In order to accomplish this, the exchange of deuterium for hydrogen at carbon 2 c an be monitored via 1 H NMR Scheme 20 : An objective of this project is to prepare mononuclear zinc and magnesium complexes in which a molecule of acetol can coordinate, in order to potentially model the enediolization step carrie d out at the active site of RuBisCo. Chapter 2: Experimental 2.1 General Considerations A Bruker AC 250 MHz NMR spectrometer was used for proton and carbon NMR experiments. Chemical shifts are relative to TMS. Melting points were obtained on a Meltemp or Meltemp II melting point apparatus Precipitates were separated by centrifugation with a Clay Adams Safeguard centrifuge. FT IR spectra were obtained on solid samples using an Avatar 320 FT IR spectrometer All reactions were run under nitrogen usin g a standard s chlenck line unless noted otherwise. A bath composed of polyethylene glycol (PEG) was used to heat reaction flasks when necessary. X ray diffraction studies were carried out by Dr. Lee Daniels of Rigaku Americas using a
48 Rigaku XtaLAB mini s ingle crystal diffractometer. Supplemental i nformation regarding data acquisition and crystallographic parameters is given in Appendix B Materials were obtained from either Sigma Aldrich or Acros Organics and used as purchased unless otherwise indicated. Dry ice was purchased from Publix and used the same day it was purchased after drying the surface by rubbing with paper towels. Dry THF was obtained from distillation over sodium and benzophenone. 2.2 Preparation of ligands 2.2 .1 B is 3,5 dimethylpyr azol 1 ylacetic acid ( H bdmpza) : Preparation of H bdmpza was adapte d from literature procedure s 51 T HF (30 mL) i s added to a flask containing 8.74 g (91.0 mmol) 3,5 dimethylpyraz ole, 1.75 g (7.5 mmol) benz yltriethylammonium chloride, 15.41 g (111.5 mmol) potassium carbonate, and 6.25 g (111.5 mmol) potassium hyd roxide. D ichloroacetic acid (2.5 mL, 30.5 mmol) is added to the suspension The suspension is heated under N 2 with vigorous stirring and refluxed for 15 h. The solvent is removed in vacuo, and water is added to the off white residue, causing some of the residue to dissolve. The aqueous solution is acidified to pH 7 with 6 M hydrochloric acid. The suspension is extracted with two washes of diethyl ether to remove unreacted pyrazole. The aqueous layer is acidified with 6 M hydrochloric acid to pH 1, and extracted with a wash of diethyl ether. A small amount of THF can be added to improve the separation between the aqueous and ether layers. The or ganic layer is dried over anhydrous sodium sulfate, gravity filtered, and concentrated in vacuo resulting in isolation of an off white powder. Recrystall ization from acetone resulted in isolation of H bdmpza as a white powder. Yield: 2.272 g (30%) mp: 15 3 154 o C (decomposed)
49 1 H NMR (CDCl 3 3 ), 2.24 (s, 6H, CH 3 ), 5.90 (s, 2H, H pz ), 6.78 (s, H, CH) 13 C NMR (CDCl 3 11.2 (CH 3 ), 13.5 (CH 3 ), 70.6 (CH), 108.0 (C pz ), 141.9 (C pz ), 149.1 (C pz ) 165 (COOH) IR: = 1743 cm 1 (C=O stretch), 1560 (C=N stretch) 2.2 .2 3,5 D iisopropylpyrazole : Preparation of 3,5 diisopropylpyrazole was adapte d from literature procedure s 61 2,6 D imethyl heptanedione (6.8 mL, 63 mmol) i s dissol ved in 40 mL ethanol. H ydrazine monohydrate (3.8 mL, 75 mmol) i s added dropwise through a septum via syringe with vigorous stirring under N 2 The solution i s refluxed for 1 hour with vigorous stirring under N 2 The solution i s washed with an aqueous sol ution of concentrated sodium chloride and extracted with diet hyl ether. All excess hydrazine in the aqueous layer is neutralized with bleach. The organic layer i s dried over anhydrous magnesium sulfate, gravity filtered, and concentrated in vacuo to isol ate 3,5 diisop r opylpyrazole as a white solid. Yield: 5.25 g (53%) mp: 89 90 o C 1 H NMR (CDCl 3 = 1.27 (d, 12H, CH 3 ), 2.96 (sep, 2H, CH), 5.89 (s, 1H, H pz ) 13 C NMR (CDCl 3 22.8 (CH 3 ), 27.1 (CH), 98.3 (C pz ), 155.1 (C pz ) IR: = 3012 3175 cm 1 (=C H stretch), 2817 2960 cm 1 (C H stretch), 1576 cm 1 (C=N stretch) 2.2 .3 B is 3,5 diisoprop ylpyrazol 1 ylmethane (bdippzm) : Preparation of 3,5 diisopropylpyrazole was adapte d from literature procedure s 51 D ichloromethane (125
50 mL) is added to a flask containing 3.16 g (20.7 mmol) 3 ,5 di isopropyl pyrazole, 0. 5 g ( 2.0 mmol) benzyltriethylammonium chloride, 1 1.25 g ( 8 1.5 mmol) anhydrous potassium carbonate, and 4. 5 g ( 80.0 mmol) potassium hydroxide. The suspension i s refluxed for 15 hours with vigorous stirring under nitrogen. Salts a re removed by filtration, and the filtrate i s concentrated in vacuo The resulting solid i s washed with water and extracted with pentane. The organic layer i s dried over anhydrous magnesium sulfate, gravity filtered, and concentrated in vacuo to isolate a y ellow oil. The oil crystallizes over a few hours to give slightly yel low crystals. The crystals are washed with water over vacuum filtration to remove diethylbenzylamine impurity resulting in slightly clearer crystals of bdippzm Yield: 2.508 g (76% ) mp: 5 2 54 o C 1 H NMR (CDCl 3 1.04 (d, 12H, CH 3 ), 1.22 (d, 12H, CH 3 ), 2.90 (sep, 2H, CH), 3.40 (sep, 2H, CH), 5.85 (s, 2H, H pz ), 6.23 (s, 2H, CH 2 ) 13 C NMR (CDCl 3 23.2 (CH 3 ), 23.5 (CH 3 ), 25.4 (CH), 28.2 (CH), 62.5 (CH), 100.1 (C pz ), 152.0 (C pz ), 158.8 (C pz ) IR: = 2868 2958 cm 1 (C H stretch), 1656 cm 1 1548 cm 1 (C=N stretch) 2.2 .4 B is 3,5 diisopropylpyrazol 1 ylacetic acid ( H bdippza) : Preparation of 3,5 diisopropylpyrazole was adapted from literature procedure s 51 62 B dippzm (1.819 g, 5.75 mmol) is dissolved in 50 mL dry THF. The clear solution is cooled to ca. 70 o C (dry ice acetone or isopropanol) with stirring un der nitrog en. N b utyllithium (5.4 mL, 8.63 mmol) in hexanes (1.6 M) is added dropwise, and the resulting cloudy yellow solution is allowed to stir under nitrogen at 70 o C for 1 h. The reaction mixture is allowed to warm slowly to 0 o C while gaseous carb on dioxide is bubbled through for 1 .5 h resulting in the
51 solution becoming gradually clearer ( Carbon dioxide gas can be e fficiently and inexpensively produce by placing dry ice in a flask that is open to a nitrogen line, and connecting the flask contain ing the dry ice to the flask containing the THF solution via rubber tubing connected to a needle Carbon dioxide gas then flows readily through the needle and bubbles into the THF solution, as long as the flask containing the THF solution is vented to atm osphere.) The solution is concentrated in vacuo resulting in isolation of a yellow residue. The residue is washed with water, resulting in solvation of some of the residue. The aqueous suspension is acidified to pH 1 with 6 M hydrochloric acid, resultin g in a white precipitate, and then extracted with two washes of diethyl ether. The organic layer is dried over anhydrous magnesium sulfate, gravity filtered, and concentrated in vacuo. T he resulting off white powder i s washed with pentane over vacuum fil tration to remove bdippzm resulting in isolation of H bdippza as a white powder Yie ld: 1.46 g (70 %) mp: 162 163 o C (decomposed) 1 H NMR (CDCl 3 3 ) 1.24 (d, 12H, CH 3 ), 2.95 (sep, 2 H, CH), 3.08 (sep, 2H, CH) 5.95 (s, 2H, H pz ), 6.97 (s, 1H, CH) 13 C NMR (CDCl 3 3 ), 22.8 (CH 3 ), 23.0 (CH 3 ) 23.6 (CH 3 ), 25.4 (CH), 27.8 (CH), 71 (CH), 101.2 (C pz ), 153.0 (C pz ), 159.1 (C pz ), 165 (COOH) IR: = 2872 2964 cm 1 (C H stretch), 1727 cm 1 (C=O stretch), 1551 cm 1 ( C=N stretch) 2.3 : Preparation of Metal Complexes 2.3 .1 Zn(bdippza) 2 : Preparation of Zn(bdippza) 2 was adapted from literature procedure s 51 M ethanol (3 mL) i s added to 0 .108 g ( 0.3 mmol ) bdippza, resulting in a suspension. Addition of 0.012 g (0.3 mmol) of sodium hydroxide results in a
52 homogeneous solution within fifteen minutes. In a separate flask, 0.055 g ( 0.1 5 mmol ) anhydrous z inc trifluoromethanesulfonate i s dissolved in 1 mL methanol. The s olution c ontaining the ligand i s transferred into the solution containing the zinc triflate. Zn(bdippza) 2 pre cipitates out as colorless micro crystals over the next hour. The precipitate i s isolated by centrifugation, followed by washing with methanol and another centrifugation. X ray suitable crystals of Zn(bdippza) 2 can be obtained from slow evaporation of dichloromethane solvent Zn(bdippza) 2 was also isolated from a 1:1 mixture of Zn(OTf) 2 and bdippza in basic methanol. Yield: 0.084 g (71% ) 1 H NMR (CDCl 3 2 H, CH 3 ), 1.13 (d, 12 H, CH 3 ), 1.26 (d, 12 H, CH 3 ), 1.40 (d, 12 H, CH 3 ), 2.58 (sep, 4H, CH), 3.08 (sep, 4H, CH), 5.90 (s, 4 H, H pz ), 6.62 (s, 2 H, CH) 13 C NMR (CDCl 3 3 ), 68 (CH), 99. 1 (C pz ), 151.3 (C pz ), 161.2 (C pz ), 166 (COO) IR: = 2867 3011 cm 1 (C H stretch), 1654 cm 1 (C=O stretch) 2.3 .2 [Zn(bdippza)Cl] 2 : M ethanol (1 mL) i s added to 0 .036 g (0.1 mmol) bdippza, resulting in a suspension. Addition of 0.004 g (0.1 mm ol) of sodiu m hydroxide results in a homogeneous solution within fifteen minutes. In a separate flask, 0.036 g (0.1 mmo l) anhydrous zinc chloride i s dissolved in 1 mL methanol. The s olution containing the ligand i s transferred into the solution containing the zinc c hloride. [Zn(bdippza)Cl] 2 immediately precipitates out of solution as a white powder The precipitate i s isolated by centrifugation, followed by washing with methanol and another centrifugation. X ray
53 suitable crystals of [Zn(bdippza)Cl] 2 were obtained from slow evaporation of chloroform or dichloromethane solvent. Yield: 0.034 g (74%) 1 H NMR (CDCl 3 1.03 (d, 6H, CH 3 ), 1.20 (d, 6H, CH 3 ), 1.30 (d, 6H, CH 3 ), 1.36 (d, 6H, CH 3 ), 3.04 (sep, 2H, CH), 3.52 (sep, 2H, CH ), 6.01 (s, 2H, H pz ), 6.60 ( s, 1H, CH) 13 C NMR (CDC l 3 22.5 (CH 3 ), 22.8 (CH 3 ), 23.2 (CH 3 ), 23.6 (CH 3 ), 26.3 (CH), 27.6 (CH), 67.4 (CH), 100.1 (C pz ), 155.1 (C pz ), 164.3 (C pz ), 166.3 (COO) IR: 3451 cm 1 (O H stretch), 2875 2962 cm 1 (C H stretch), 1688 cm 1 (C=O stret ch) 2.3 .3 Mg(bdippza) 2 : Methanol (1 mL) is added to 0.036 g (0.1 mmol) Hbdippza, resulting in a suspension Addition of one equivalent of sodium hydroxide results in a homogeneous solution within fifteen minutes. In a separate flask, 0.036 g ( 0.1 1 mmol ) anhydrous magnes ium trifluoromethanesulfonate i s dissolved in 1 mL methanol. The s olution containing the bdippza ligand i s transferred into the solution containing the zinc triflate. Mg(bdippza) 2 precipitates out of methanol as colorless x ray quality crystals over a few days Alternatively, Mg(bdippza) 2 can be prepared under analogous conditions using MgCl 2 as a magnesium source. Alternative preparation of Mg(bdippza) 2 : Hbdippza (0.108 g, 0.3 mmol) is dissolved in 10 mL dry THF. In another flask, 0.7 mL (0.35 mmol) 0.5 M di n butylmagnesium in heptanes is dissolved in 10 mL dry THF. The solution containing the Hbdippza is added dropwise to the magnesium solution at 78 o C with stirring. The solution is stirred at 78 o C for 2 hours, and then allowed to warm to room temperature. The solution becomes
54 cloudy as it warms. Solvent is removed in vacuo, resulting in isolation of a mixture of Mg (bdippza) 2 and di n butylmagnesium by proton NMR. 1 H NMR (CDCl 3 0.94 (d, 6H, CH 3 ), 1.12 (d, 6H, CH 3 ), 1.27 (d, 6H, CH 3 ), 1.40 (d, 6H, CH 3 ), 2.60 (sep, 2H, CH), 3.08 (sep, 2H, CH), 5.88 (s, 2H, H pz ), 6.61 (s, 1H, CH) 13 C NMR (CDCl 3 22.0, 22.2, 24.1, 24.8, 26.1, 27.2 (CH/CH 3 ), 68.5 (CH), 99.1 (C pz ), 151.7 (C pz ), 161.9 (C pz ) IR: 2850 29 62 cm 1 (C H stretch), 1664 cm 1 (C=O stretch) 2.3 .4 Zn(bdmpza) 2 : Preparation of Zn(bdmpza) 2 was adapted from literature procedure s 51 M ethanol (2 mL) is added to 0 123 g ( 0. 5 mmol) bdippza, resulting i n a suspension. Addition of 0.0 2 0 g ( 0 .5 mm ol) of sodium hydroxide results in a homogeneous solution within fifteen minutes. In a separate flask, 0.182 g ( 0.5 mmol) anhydrous zi nc trifluoromethanesulfonate is dissolved in 1 mL methanol. The s olution con taining the ligand i s transferred into the solution containing the zinc triflate. Zn(bdm pza) 2 precipitates out as colorless crystals over the next hour. The precipitate i s isolated by centrifugation, followed by washing with methanol and another centrifu gation. Alternatively, Zn(bdmpza) 2 can be prepared under analogous conditions using ZnCl 2 as a zinc source. Yield: 0.252 g (90 %) 1 H NMR (CDCl 3 1.97 (s, 6H, CH 3 ), 2.45 (s, 6H, CH 3 ), 5.87 (s, 2H, H pz ), 6.56 (s, 1H, CH)
55 13 C NMR (CDCl 3 11.3 (CH 3 ), 13.1 (CH 3 ), 67.3 (CH), 107.1 (C pz ), 140.2 (C pz ), 150.6 (C pz ), 166.9 (COO) IR: 1647 cm 1 (C=O stretch) Chapter 3: Results and Discussion 3.1 Ligand Synthesis: Analysis of Reaction Mechanisms and Spectra 3.1.1 H bdmpza: In order to gain experience with the reactions involved in preparing Hbdippza, bis 3,5 dimethylpyrazolylacetic acid (Hbdmpza) was prepared using the synt hesis Scheme 11 ) Although it had been found that s synthesis could not be applied to the sterically hindered 3,5 di tert butylpyrazole, it had not yet been attempted for the less hindered 3,5 diisopropylpyrazole. Using this procedure, bdmpza is prepared from the reaction of 3,5 dimethylpyrazole with either dibromoacetic acid or dichloroacetic acid. Although a higher yield can be obtained using dibromoacetic acid, dichloroac etic acid was used because it is significantly less expensive. Thus, the ligand was synthesized from the commercially available reagents dichloroacetic acid and 3,5 dimethylpyrazole. The reaction also requires hydroxide ions, which deprotonate the pyrazo le nitrogen either before or after a pyrazole nitrogen initiates nucleophilic attack to replace the chlorides of the dichloroacetic acid. After one chlori de has been replaced, the second chloride is somewhat more easily displaced because the pyrazole nitr ogen is a better donor than the chloride, putting more electron density on the other chloride leaving group. Although potassium hydroxide is normally insoluble in tetrahydrofuran, the organic triethylbenzylammonium ( TEBA ) cation, which is sparingly solubl e in THF, acts as a
56 solid liquid phase transfer catalyst, transporting hydroxide ions into solution. Because of the phase transfer conditions necessary for carrying out the reaction, the reaction proceeds slowly and gives relatively low yield After the reaction was carried out, the solvent was removed to isolate a mixture of mostly the potassium salt of 3,5 dimethylpyrazol 1 ylacetate, potassium hydroxide, potassium carbonate, TEBA, and s tarting material. The product wa s isolated from starting material by extracting it from ether solution into an aqueous la yer at pH 7, at which point it i s still deprotonated and therefore more so luble in water than ether. It wa s then further separated from phase transfer catalyst and potassium salts by acidification to pH 1 protonating the potassium salt to form Hbdmpza and extraction from the aqueous layer into diethyl ether. After the eth er had been removed, the acid was further purified by recrystallization from acetone. A yield of approximately 30% was isolated ea ch time this reaction was carried out. Yields of up to 45% are reported using dibromoacetic acid. 51 HNMR, CNMR, and FT IR spectra, as well as melting point information provide evidence that the product is pure bdmpza (reported in experimental, section 2.2.1) 3.1.2 3,5 diisopropylpyrazole: Unlike 3,5 dimethylpyrazole, 3,5 diisopropylpyrazole is not available commercially. Therefore, t he first step in the synthesis of the ligand bdippza is preparation of 3,5 diisopropylpyrazol e ( Scheme 21 ) The synthesis wa s carried out from commercially available starting materials hydrazine mono hydrate and 2,6 dimethyl 3,5 heptanedione using dry ethanol as solvent A predi cted reaction mechanism is shown in Scheme 22 Proton transfer steps are omitted. The reaction is initiated by nucleophilic attack of a hydrazine nitrogen at either of the two equivalent carbonyl carbons. After a proton transfer mediated by solvent molecules, the other
57 nitrogen is able to attack the other carbonyl carbon to form a stable 5 membered ring structure. Since the reaction wa s run under basic conditions, the next step is likely a base catalyzed E2 elim ination of one of the hydroxides to form an enamine species. Another base catalyzed elimination then becomes very energetically favorable due to formation of an aromatic pyrazole ring. Scheme 21 : Preparation of 3,5 diisopropyl pyrazole from 1,6 dimethyl 3,5 heptanedione and hydrazine hydrate After re fluxin g, the product of the reaction wa s isolated from excess hydrazine by extraction from an aqueous solution of concentrated sodium chloride into diethyl ether. The excess hydra zine remained in the aqueous layer and the pyrazole product wa s extracted into the organic phase. After rotovapping to remove ethe r and ethanol, a white powder was isolated that was effectively pure 3,5 diisopropylpyrazole, according to its NMR spectra. The proton NMR spectrum ( Figure 41 in Appendix A ) showed a sharp doublet at 1.27 ppm integrating for six protons corresponding to the methyl protons of the isopropyl moieties. A septet at 2.96 ppm integrating for two protons corresponds to the isopropyl methine protons. Finally, a slightly broader singlet at 5.89 ppm that integrates for one proton corresponds to the proton bound to the pyrazolyl carbon. The carbon NMR spectrum ( Figure 42 in Appendix A ) showed two resonances corresponding to pyrazolyl carbons at 98.3 and 155.1 ppm and two resonances corresponding to isopropyl carbons at 22.8 and 27.1 ppm. The fact that only one set of resonances is seen for the
58 isopropyl g roups indicates that they are in chemically equivalent environments. This means that the proton bound to the pyrazolyl nitrogen exchanges from one nitrogen to another at a rate faster than the NMR timescale, so that the NMR spectrum shows an average reson ance instead of two separate resonances. The N H peak cannot actually be seen in the proton NMR spectrum likely due to quick relaxation caused by coupling to quadrupole Scheme 22 : Proposed reaction mechani sm for conversion of 1,6 dimethyl 3,5 heptanedione to 3,5 diisopropylpyrazole Proton transfer steps are omitted. 3. 1.3 bdip pzm : dichloroacetic acid was attempted several times under various conditions. Although it
59 was known that the considerable steric hindrance present in 3,5 di tert butylpyrazole prevented this reaction from occurring under the same conditions as for 3,5 dimethylpyrazole, it had not yet been attempted for diisopro pylpyrazole. When the reaction was attempted using the same conditions as in the preparation of bdmpza, only starting material was isolated after refluxing. Assuming that a higher activation energy might be required to drive the reaction, a higher boilin g solvent was used for the reaction so that the solution could be refluxed at higher temperature. However, when dimethoxyethane (DME) was used in place of THF, starting material was once again isolated after refluxing. The phase transfer catalyst TEBA wa s substituted by tetrabutylammonium iodide in the hopes that iodide would substitute for chloride in dichloroacetic acid in situ, creating a better leaving group fo r the pyrazole to substitute This approach resulted in isolation of starting material comb ined with unknown degraded products, possibly resulting from radical processes, but no 3,5 diisopropylpyrazol 1 ylacetic acid. Bdippza was thus su bsequently prepared by the 2 step method initially used by Otero. In order to synthesize H bdippza bdip pzm wa s prepared from diisopropylpyrazole and isolated as an intermediate step. B dip pzm was produced via t he reaction shown in Scheme 23 As in the synthesis of bdmpza, bdippzm is prepared under solid liquid phase tran sfer conditions. In this case, however, the dichloromethane reagent was also used as solvent, greatly simplifying the synthesis. Although dichloromethane boils at 40 o C, so that reflux is carried out at relatively low temperature, the pyrazole reagent is fully converted to product within a few hours because a very large excess of dichloromethane (DCM) is used (roughly 100 molecules of DCM to 1 molecule of pyrazole).
60 Scheme 23 : Preparation of bdippzm from 3,5 diisopropylpyrazole and dichloromethane A plausible mechanism for the reaction is shown in Scheme 24 Assuming the reaction follows an S N 2 mechanism, the rate of the reaction is likely first order in both DCM and diisopropylpyrazole Since DCM is also the solvent, its concentration remains effectively constant, and very large relative to the concentration of pyrazole, throughout the reaction The reaction rate can therefore be considered pseudo first order in diisopropylpyrazole. An excess of hydroxide base is used to deprotonate the pyrazole, although deprotonation may not be necessary for a pyrazole nitrogen to attack dichloromethane. Another plausible mechanism involves deprotonation prior to n ucleophilic attack to form an a nio nic pyrazole species, although the protonated pyrazole is likely a good enough nucleophile to attack dichloromethane before deprotonation The potassium hydroxide base may be necessary to deprotonate pyrazole after nucleophilic attack to form a neutral sp ecies. The purpose of the potassium carbonate, used in equimolar amounts with potassium hydroxide, is to scavenge water molecules formed by deprotonation of the pyrazole ring, so that water cannot potentially hydrolyze the C N bonds in the product. A cat alytic amount of TEBA was again used as a phase transfer catalyst, transporting hydroxide ions into the dichloromethane solution.
61 Scheme 24 : Proposed reaction mechanism for conversion of 3,5 diisopropylpyrazole to bdippzm The re action was followed by proton NMR spectroscopy to determine when the 3,5 diisopropylpyrazole had been completely converted to bdippzm. Because the product bdippzm was the only neutral organic species predicted to be present in solution under the basic con ditions present, it was isolated simply by extraction from aqueous solution into pentane after evapo rating excess dichloromethane. A proton NMR spectrum ( Figure 43 in Appendix A ) again showed complete conversion t o products, as evidenced by two doublets at 1.04 and 1.22 ppm integrating for twelve protons each, corresponding to the
62 two sets of distinct methyl groups, two septets at 2.90 and 3.40 ppm integrating for two protons each, corresponding to the two inequiva lent isopropyl methine groups, a singlet at 5.85 ppm integrating for two protons, corresponding to the two equivalent pyrazole protons, and a singlet at 6.23 ppm integrating for two protons, corresponding to the central methylene group. A carbon NMR spect rum ( Figure 44 in Appendix A ) showed four resonances from 23 to 28 ppm corresponding to the four distinct isopropyl carbons, one resonance at 62.5 ppm corresponding to the methylene carbon, and resonances at 100.1, 152.0, and 158.8 ppm, corresponding to the pyrazolyl carbons. On the initial proton NMR spectrum taken after extraction with pentane ( Figure 29 ) three additional resonances, a singlet at 3.57 ppm, a quartet at 2.52 ppm and a multiplet at 7.33 ppm were found in the spectrum. These resonances were much less intense than the bdippzm resonances, but close in intensity to one another giving evidence that they all arose from the same contaminant. It was found that the peak positions, intensities, and multiplicities correspond very closely to the literature values for the molecule diethylbenzylamine (DEBA), a neutral amine produced by removal of one of the ethyl groups of the phase transfer catalyst TEBA. The resona nce corresponding to the methyl groups of DEBA coincides with the resonance corresponding to one of the bdippzm methyls at 1.04 ppm, and is therefore not visible in the spectrum.
63 Figure 29 : 1 H NMR spectrum of bdmpzm with DEBA impurity in CDCl 3 Assignm ents of peaks are labeled. The unlabeled peaks correspond to TMS, vacuum grease, H 2 O, and CHCl 3 going from right to left. The D EBA may have been produced by nucleophilic substitution of hydroxide for DEBA, re sulting in formation of ethanol, but because TEBA is sterically hindered, it is more likely that an elimination occurred, producing ethylene gas ( Scheme 25 ). N o triethylamine was observed to be produced in the r eaction, which would be expected if 1 1, 1 4 3 3 4 1 1 2
64 the benzyl group experienced nucleophilic attack ; h owever, unlike DEBA, triethylamine has a relatively high vapor pressure and would likely be removed during evaporation of solvent U nlike the ethyl groups, a benzyl gro up does not have the ability to undergo an elimination reaction to eliminate triethylamine because the aromatic carbon adjacent to the methylene does not have any protons. Scheme 25 : A) S N 2 mechanism for conversion of TEBA to DEBA B) E2 mechanism for conversion of TEBA to DEBA Regardless of how it was formed, it was necessary to remove the DEBA impurity in order to proceed to the next step in preparing the ligand. Fortunately, diethylbenzylamine has a pKa of approximatel y 9.5, and is therefore readily protonated and solubilized in neutral water. Washing with deionized water over vacuum filtration was therefore sufficient to remove DEBA, resulting in pure bdippzm by proton NMR spectroscopy The crystals were found to mel t at around 5 3 o C, which is much lower than the temperature reported in the literature for the similar compound bdtbpzm, which melts at 132 o C 51 It is possible that the low melting point resulted due to initial crystallization in the presence of A B
65 DEBA impurity. Upon removal of the DEBA, there may have been major def ects in the crystal lattice that caused the compound to melt relatively easily. A f t e r o n c e a g a i n s o l i d i f y i n g h o w e v e r t h e s o l i d w a s s t i l l f o u n d t o m e l t a t t h e s a m e t e m p e r a t u r e i n d i c a t i n g t h a t t h e l o w m e l t i n g p o i n t m a y b e m o r e l i k e l y d u e t o t h e p r e s e n c e o f s o m e i m p u r i t i e s The crystals were, however, of sufficient purity to p roceed forward to the next step; carboxylation of the central methylene carbon to form Hbdippza. 3.1.4 H bdippza: H bdippza was synthesized by deprotonation of bdippzm with n butyllithium, followed by carboxylation with carbon dioxide and acid workup ( Scheme 26 ). Scheme 26 : Preparation of Hbdippza from bdmpza and carbon dioxide n B utyllithium is a very strong base capable of deproto nating bdippzm to form the reactive organometallic species bis 3,5 diisopropylpyrazol 1 ylmethyllithium (Libdippzm) n B utyllithium was added slowly at 78 o C because it is known to readily deprotonate the solvent, THF, at room temperature. Even with the possibility of reaction, THF is a particularly good solvent fo r deprotonations using BuLi because it coordinate s to lithium, making the butyl carbanion more reactive. The reaction mechanism for conversion of bdippzm to Hbdippza is shown in Scheme 27 After formation of the
66 l ithiated species, the solution wa s exposed to gaseous CO 2 generated from dry ice and allowed to rise from dry ice temperature to 0 o C. Nucleophilic attack at the carbon in CO 2 result ed in the lithium salt of the carboxylated ligand bdippza. After the re action was completed and the THF solvent was removed, the product was treated with acid to produce the carboxyli c acid Hbdippza. The product was then isolated by extraction from aqueous solution into diethyl ether. Evaporation of the ether results in a p owder that is mostly composed of bdippza, but also contains some bdippzm starting material by proton NMR spectroscopy Since a large excess of n butyllithium reagent was used, the isolated bdippzm is likely the result of reaction of the deprotonated speci es with water, rather than the deprotonation not occurring in the first place. Although the experimental setup is specifically designed with the intention of avoiding water, with all glassware being thoroughly oven dried, some exposure to water cannot be avoided. The main sources of water are likely the dry ice from which the carbon dioxide is prod uced, which was purchased from P ublix and briefly dried with a paper towel before use, and the tubing, which could not be oven dried The first time this react ion was carried out successfully, nearly half of the isolated solid was starting material. However, the ratio of product to starting material isolated the next time the reaction was carried out was approximately 10:1. The better product yield could likel y be attributed to improved technique in eliminating moisture and to the larger scale of the reaction, which made small, unavoidable sources of moisture less important. Fortunately, bdippza could be easily separated from bdippzm by washing with pentane, i n which the polar, protic acid product is nearly completely insoluble, while the largely nonpolar starting reagent is very soluble.
67 Scheme 27 : Proposed reaction mechanism for conversion of bdippzm to bdip pza A proton NMR spectru m of the purified bdippza ( Figure 30 and Figure 45 in Appendix A ) showed a multiplet centered at 0.99 ppm and a doublet at 1.24 ppm each integrating for twelve protons, cor responding to the methyl groups, septets at 2.95 and 3.08 ppm integrating for four protons in total (the two septets could not be completely resolved), correspondin g to the isopropyl methines, a singlet at 5.95 integrating for two protons corresponding to the pyrazolyl proton and a singlet at 7.01 ppm integrating for one proton, corresponding to the cen tral methine proton A resonance corresponding to the carboxylate proton was not seen in the spectrum, likely due to either intra or intermolecular hydro gen bonding. An unexpected feature of this NMR spectrum is that the resonance corresponding to one set of methyl protons at 0.99 ppm shows up as a multiplet composed of two overlapping doublets, rather than a single doublet. Interestingly, this effect ha s not been seen in the spectra for either H bdmpza or H bdtbpza. In several spectra collected, there are four distinct peaks centered at 0.99 ppm with roughly equivalent intensities, but some other spectra show just three peaks that appear similar to a trip let because the two central peaks have broadened into one another. This may be the result of differe nces in shimming between different acquisitions of spectra
68 Figure 30 : 1 H NMR spectrum of Hbdippza in CDCl 3 A ssignments of peaks are labeled. The unexpected splitting pattern can be rationalized by assuming the compound experiences some form of hindered rotation, so that either the two methyls on the same isopropyl moiety are inequivalent, or the methyls on one pyrazole are inequivalent to the analogous methyls on the other pyrazole. Methyl groups in the two slightly different environments cannot interconvert on the NMR timescale, resulting in two doublets rather 1 2 3 4 4 3 1
69 than one. From the NMR spectrum, it is clear tha t only one set of methyl groups experience this effect, since no unexpected splitting appears in any of the other resonances. The carboxylate group likely introduces enough steric hindrance that the isopropyl groups cannot freely rotate. It is uncertain which isopropyl groups are subject to the additional splitting. In theory, the four peaks should merge back into two peaks at higher temperature since the rotation barrier should be overcome given enough energy. However, it is possible that the compound would decompose before attaining a temperature sufficient to overcome this barrier. 2 D NMR experiments might also provide more insight into the origins of the unexpected splitting; however, neither of these experiments has been carried out. A carbon NMR spectrum ( Figure 31 and Figure 46 in Appendix A ) showed six peaks between 22.7 and 27.8 ppm corresponding to the isopropyl groups, a peak at 71 ppm corresponding to the cen tral methine carbon, peaks at 101.2, 153.0, and 159.1 ppm corresponding to the pyrazolyl carbons, and a peak at 166 ppm corresponding to the carboxylate c arbon. This spectrum also shows evidence of hindered rotation. While only four resonances are expect ed for the isopropyl groups, there are clearly six present in the spectrum. This is consistent with the splitting seen in the proton NMR spectrum, but does not give much more information as to its origin. An FT IR spectrum ( Figure 47 ) showed a carbonyl stretch at 1727 cm 1 Hbdippza was thus prepared in 28% overall yield in three steps. It could then be deprotonated and utilized in carboxylate form as a tridentate ligand.
70 Figure 31 : 13 C NMR spectrum of Hbdippza in CDCl 3 Assignments of peaks are labeled. 3 .1.5 Other Organic Syntheses : Initially, synthesis of 2,6 dimethyl 3,5 heptanedione was attempted from the starting materials isobutyryl chloride and 3 methy l 2 butanone under basic conditions. This resulted in an intractable mixture of products containing little, if any, of the desired product, possibly resulting partly from deprotonation of 3 1 8 2 3 7 5 4 6 1 4 8 5 6 7 3 2
71 methyl 2 butanone at multiple sites. This reaction was not pursu ed, and the starting reagent 2,6 dimethyl 3,5 heptanedione was instead purchased from Acros Organics. In order to practice procedures using less expensive reagents, dimethylpyrazole, bis 3,5 dimethylpyrazol 1 ylmethane (bdmpzm) and bdmpza were prepared. Dimethylpyrazole was synthesized from acetylacetone and hydrazine via the same method as diisopropylpyrazole ( Scheme 21 ) Bdmpzm was synthesized from dimethylpyrazole and dichloromethane via the same method as fo r bdippza ( Scheme 23 ) ; however workup was slightly changed. Due to the insolubility of bdmpzm in pentane, bdmpzm was extracted with diethyl ether instead. Finally, H bdmpza was prepared from bdmpzm n butyllithium and carbon dioxide via the same method as for H bdippza ( Scheme 26 ) All syntheses resulted in lower product yields than later syntheses involving isopropyl derivatives in part due to less experience with the proc edures. 3.2 Metal Complexes: Synthesis, Spectra, and Crystallographic Data 3.2.1 Zinc complexes containing bdippza: Even at greater than one to one metal to ligand ratio, only the bis ligand complex Zn(bdippza) 2 was isolated from the reaction of Hbdippza with Zn(OTf) 2 in basic methanol ( Scheme 28 ). Scheme 28 : Preparation of Zn(bdippza) 2 from Hbdippza and zinc triflate in basic methanol
72 Several factors ga ve strong evidence that it is th e bis ligand complex, rather th an a mono ligand complex, that wa s isolated. T he proton NMR spectrum ( Figure 32 and Figure 48 in Appendix A ) showed that the resonance corre sponding to the central methine proton s moves upfield from that of the free ligand. Additionally, the carbon NMR spectrum ( Figure 49 ) showed that the resonance corresponding to the central carbon moves upfield in the complex relative to the free ligand Although this is somewhat counterintuitive, since binding to an electrophilic metal center generally inductively decreases the electron density at a ligand, t his is characteristic of formation of a bis ligand compl ex, and is also seen in Zn(bdmpza) 2 51 Conversely, formation of a mono ligand, tetrahedral complex containing a more weakly donating group, such as triflate or a halogen at the fourth site would likely result in a downfield shift in these peaks as in the mono ligand complex (bdtbpza)ZnCl reported in the same paper A simple explanation for this difference is that the binding of two bdippza ligands increases the overall amount of electron donation to th e metal, so that less electron density is drawn away from either ligand individually. This causes greater electron density to reside at the central carbon, shielding it relative to the free ligand. Another, complementary explanation can be formulated by making use of the 18 electron rule. If both bdippza ligands were bound as six electron donors, the zinc complex would have a total of twenty two valence electrons (ten from zinc, six from each ligand), a number that is unrealistic for a stable zinc specie s. It is much more likely that the structure of Zn(bdippza) 2 is th e result of a resonance hybrid in which only two of the four pyrazolyl nitrogens are donating to the metal. In this resonance hybrid, the pyrazolyl nitrogens can essentially be considered as one electron donors. In (bdtbpza)ZnCl and theoretically its isopropyl counterpart, the complex is an 18 electron
73 species with a weakly donating chloride ligand bound so that significantly more electron density is required from the bispyrazolylacetate ligand relative to the bis ligand complex. Figure 32 : 1 H NMR spectrum of Zn(bdippza) 2 in CDCl 3 A more conclusive piece of evidence as to the identity of the isolated complex is that the carbon NMR spectrum does not show a re sonance corresponding to a triflate carbon. This resonance would show up as a quartet at around 120 ppm due to coupling to the spin 19 F nuclei if a triflate anion were present in the compound. Additionally, the proton
74 NMR, carbon NMR and the FT IR spec trum do not show evidence of a methoxide or hydroxide ligand, which could theoretically occupy the fourth coordination site in a neutral tetrahedral complex. Thus, it is most likely that another bdippza ligand is coordinated rather than another anionic sp ecies. An X ray crystal structure of the compound was obtained ( Figure 33 ), which confirms the structure of Zn(bdippza) 2 The proton NMR spectrum of Zn(bdippza) 2 ( Figure 32 and Figure 48 in Appendix A ) shows four doublets between 0.94 and 1.40 ppm each integrating for twelve protons. These resonances correspond to methyl protons in four different electronic environments. This is a n additional splitting from the free ligand, which shows three doublets. No such splitting is seen in either Zn(bdmpza) 2 or Zn (bdtbpza) Cl. 51 This additional doublet likely arises due to extra hindered ro tation in the isopropyl groups in the metal complex relative to the free ligand. The rest of the pea ks in the NMR spectrum have shifts and intensities similar to what would be expected. Two septets at 2.55 and 3.08 ppm correspond to the isopropyl methine s, and singlets at 6.03 and 6.62 ppm correspond to the pyrazolyl proton and central methine proton, respectively. The carbon NMR spectrum ( Figure 49 in Appendix A ) shows six resonances between 21.9 ppm and 26.8 pp m, two of which correspond to isopropyl methine carbons and four of which correspond to methyl carbons. The existence of four separate methyl peaks is in agreement with the proton NMR spectrum indicating four distinct methyl groups Again, the rest of th e spectrum shows up as expected; the methine carbon resonance is at 68 ppm, the pyrazolyl resonances are at 99.1, 151.3, and 161.2 ppm, and the carboxylate resonance is at 166 ppm. The FTIR spectrum ( Figure 50 ) sh ows a carbonyl stretch at 1654 cm 1 which is within the expected range for a carboxylate group. T his peak has moved to significantly
75 lower wavenumber from the free ligand, which shows a carbonyl stretch centered at 1727 cm 1 Binding to zinc therefore c learly weakens the C=O bond relative to the protonated form of the ligand. An ORTEP diagram of Zn(bdippza) 2 is shown in Figure 33 Structure determination details are summarized in Table 1 Further experimental details and a table of crystallographic parameters are giving in Appendix B. Figure 33 : ORTEP diagram of Zn(bdippza) 2 Atoms in the asymmetric unit are labeled. The x ray crystal struct ure of Zn(bdippza) 2 shows that the methyl groups of the pyrazole rings are directed away from the zinc, as is also the case for hindered octahedral MTp iPr2 complexes. 2 The molecule has an inversion cente r at the zinc atom, so that only one bound ligand is in the asymmetric unit. The compound can be classified in the C 2h point group.
76 Empirical Formula C40H6 2 N8O4Zn Formula Weight 782.35 Crystal Color, Habit colorless, prism Crystal Dimensions 0.200 X 0.190 X 0.180 mm Crystal System M onoclinic Lattice Type Primitive Lattice Parameters a = 10.166(4) b = 17.284(6) c = 12.528(4) = 96.866(7) o V = 2185(2) 3 Space Group P21/c (#14) Z value 2 D calc 1.189 g/cm 3 F 000 836 (MoK ) 6.0 82 cm 1 Temperature 20.0 o C Reflections Measured Total: 15227 Unique: 4997 Structure Solution Direct Methods (SHELX97) Refinement Full matrix least squares on F2 Function Minimized w (Fo2 Fc2)2 Least Squares Weights w = 1/ [ 2(Fo2) + (0.0422 P)2 + 0.6940 P ] where P = (Max(Fo2,0) + 2Fc2)/3 2 max cutoff 55.0 o Anomalous Dispersion All non hydrogen atoms No. Observations (All reflections) 4997 No. Variables 249 Reflection/Parameter Ratio 20.07 Residuals: R1 (I>2.00 (I)) 0.0398 Residuals: R (All reflections) 0.06 Residuals: wR2 (All reflections) 0.0991 Goodness of Fit Indicator 1.016 Max Shift/Error in Final Cycle < 0.001 Max peak in Final Diff. Map 0.23 e / 3 Min peak in Final Diff. Map 0.26 e / 3 Table 1 : Data acquisition and structure determination details for Zn(bdippza) 2
77 Selected bond lengths and angles are reported in Table 2 in comparison to bond lengths and angles for Zn(bdmpza) 2 ( Figure 21 ) reported by Burzlaff and coworkers. 51 It can be noted that Zn(bdippza) 2 has a Zn O bond length that is 0.08 5 (3) shorter than Zn(bdmpza) 2 This is the most significan t difference in two otherwise very similarly structured compounds. The shorter distance is likely primarily due to the absence of water molecules in the crystal lattice. It was found that Zn(bdmpza) 2 crystallized with three molecules of water per one mol ecule of Zn(bdmpza) 2 while Zn(bdippza) 2 crystallized without any solvent in the crystal lattice. Incorporation of water into the crystal lattice of Zn(bdmpza) 2 likely occurred because Zn(bdmpza) 2 was crystallized from a 20:1 MeOH to H 2 O solution, and bec ause the less bulky methyl groups make bdmpza less hydrophobic than the bulkier bdippza groups in bdippza. Hydrogen bonding of a water molecule to the carbonyl oxygen in Zn(bdmpza) 2 likely weakens the Zn O bond relative to the Zn O bond in Zn(bdippza) 2 The bond angles O1 Zn1 N1, O1 Zn1 N4, and N1 Zn1 N4 in Zn(bdippza) 2 are all slightly less than 90 o as is common in six coordinate complexes containing scorpionate ligands, because the ligand does not have enough flexibility to form a complex that is close r to ideal octahedral geometry. Hence, the geometry of Zn(bdippza) 2 and of most bis scorpionate metal complexes, can be considered distorted octahedral. Distance () Zn(bdippza) 2 Zn(bdmpza) 2 Angle ( o ) Zn(bdippza) 2 Zn(bdmpza) 2 Zn1 O1 2.034 (1) 2.119 (3) O1 Zn1 N1 86.44 (6) 86.4 (1) Zn1 N1 2.183 (2) 2.180 (3) O1 Zn1 N4 86.50 (6) 86 .1 (1) Zn1 N4 2.170 (2) 2.178 (3) N1 Zn1 N4 82.40 (6) 84.4 (1) Table 2 : Selected bond distances and angles of Zn(bdippza) 2 shown in comparison to selec ted bond distances and angles of Zn(bdmpza) 2 obtained from ref. 51
78 Table 3 As reported in Table 3, the dihedral angles defined by C7 C8 C10 C9, C16 C17 C19 C20, a nd C16 C15 C13 C12 are all close to 90 o while the dihedral angles defined by C7 C8 C10 C 11 C16 C17 C19 C18 and C16 C15 C13 C14 are all close to 30 o The dihedral angles C7 C6 C4 C3 and C7 C6 C4 C5, are 74.6 o and 49.1 o respectively, which are further from 90 o and 30 o than is seen for the other isopropyl groups, likely due to steric interactions with the isopropyl groups of the other bound bdippza ligand. rings while C11, C5, C18, and C14 c Dihedral Angle ( o ) C7 C8 C10 C9 27.5 (3) C7 C6 C4 C5 49.1 (3) C16 C17 C19 C18 27.3 (3) C16 C15 C13 C1 4 28.6 (4) C7 C8 C10 C11 95.5 (3) C7 C6 C4 C3 74.6 (3) C16 C17 C19 C20 96.4 (3) C1 6 C15 C13 C1 2 94.9 (3) Table 3 : Dihedral angles of methyl groups with respect to the pyrazolyl planes The equatorial and axial methyl groups are in slightly different chemical environments from one another, which may give rise t o the 1 H NMR splitting pattern for the resonances corresponding to the methyl groups, which contains four distinct doublets. Due to the steric constraints that the isopropyl groups enforce on one another, axial and equatorial methyls may be unable to inte rconvert in solution on the NMR timescale. An other interesting feature to note regarding the solid state structure of Zn(bdippza) 2 is that t he carbonyl oxygen is in very close proximity to the central methine proton and two isopropyl methine protons of an other molecule of Zn(bdippza) 2 ( Table 4 ) This
79 interaction is not strong enough to be considered hydrogen bonding, but does likely arise due to the stabilization that results from placing the partial positive char ge s of the hydrogen atoms near the partial negative c harge of the oxygen atom in the crystal lattice. Interaction Distance () O2 2.425 O2 2.310 O2 2.412 Table 4 : Distances between O2 and hydrogen atoms from anoth er molecule of Zn(bdippza) 2 within the crystal lattice. Hydrogen atoms are labeled with the number of the carbon atom to which they are bonded. Because triflate is a weakly coordinating anion, it is easily displaced by a bdippza ligand. Binding of one b dippza molecule likely activates the resulting complex toward binding of another molecule of bdippza, so that only the bis ligand complex is formed, rather than the mono ligand complex or a mixture of both. An anion that is more difficult to displace may favor formation of a mono ligand complex. Th erefore, an analogous reaction was carried out by combining ZnCl 2 and bdippza in a one to one ratio in basic methanol ( Scheme 29 ) Scheme 29 : Preparation of [Zn(bdippza)Cl] 2 from Hbdippza and zinc c h l o r i d e in basic methanol
80 Based on the proton NMR carbon N MR and FT IR spectra, this resulted in the isolation of a different species than from the reaction of Zn(OTf) 2 with bdip pza; although some Zn(bdippza) 2 was present when the ratio of Hbdippza to ZnCl 2 was greater than one to one The proton NMR spectrum ( Figure 34 and Figure 51 in Appendix A ) again show ed four distinct doublets, each integrating for six methyl protons between 1.01 and 1. 38 ppm. The spectrum also showed two resonances correspond ing to isopropyl methines at 3.04 and 3.52 ppm, a resonance corresponding to the pyrazolyl proton at 6.01 ppm, and a resonance corresponding to the central methine proton at 6.60 pp m. The carbon NMR spectrum ( Figure 52 in Appendix A ) showed six resonances between 22.5 ppm and 27.6 ppm, corresponding to the isopropyl carbons, a resonance at 67.5 ppm corresponding to the central methine carbon, resonances at 100.1, 155.1, and 164.3 corresponding to the pyrazolyl carb ons, and a resonance at 166.3 corresponding to the carboxylate carbon. T his complex is clearly different from Zn(bdippza) 2 suggest ing that only one molecule of bdippza is bound to the zinc center, as would be desired for a complex that would be useful in enzyme model studies. A fourth coordination site could potentially be occupied by a chloride ligand, but also could have been occupied by a hydroxide or aqua ligand Another possi bility is that the complex is 5 coordinate with bridging hydroxide ligand s a structure that is not uncommon in zinc complexes. 18 The FT IR ( Figure 53 ) shows a carbonyl stretching frequency centered at 1688 cm 1 indicating a stronger C=O bond than in the bis ligand complex, but a weaker bond than in the free ligand. This is consistent with a structure in which only one bdippza is bound to a zinc center, since a ligand such as chloride or hydroxide would likely be less electron donatin g than another molecule of
81 bdippza. R e c e n t l y t he structure of the isolated compound has been confirmed as [Zn(bdippza)Cl] 2 via single crystal x ray diffraction Figure 34 : 1 H NMR spectrum of the product isolated from the reaction of bdippza and ZnCl 2 taken in CDCl 3 Good quality crystals of [Zn(bdippza)Cl] 2 were grown from dichloromethane, and were characterized with single crystal X ray diffraction. The structure shown in Figure 35 is
82 only a preliminary structure, as the x ray data was taken very recently, and has not yet been fully refined. Thus, bond le ngths and angles are not reported herei n. Figure 35 : Preliminary x ray crystallographically determined structure of [Zn(bdippza)Cl] 2 Atoms are represented as spheres (grey carbon, blue nitrogen, red oxygen, maroon zinc, yellow chlorine). Hydrogen atoms are omitted for clarity. The structure is a carboxylate bridged dimer in the solid state, but does in fact contain a one to one ratio of ligand to zinc. It is highly possible that the compound exists as a monomer in solution, and crystallizes as a dimer. Therefore, the compound may be useful as a synthetic model for zinc enzymes. The structurally similar compound [Zn(bpa tBu2,Me2 )Cl] 2 ( Figure 36 ) has been reported by Burzlaf 3 Isolation of these analogous compounds indicates bpa tBu2,Me2 and bdippza likely have very similar steric profiles. As is also the case for bdippza, the bis ligand complex Zn(bpa tBu2,Me2 ) 2 co uld be isolated under some conditions.
83 Figure 36 : Structure of Zn(bpa tBu2,Me2 ) 2 reported by Burzlaff (Taken from ref. 3 ) 3.2.2 Magnesium complexes containing bdippza: In order to synthetically model the active site of RuBisCo, we sought mono ligand magnesium complexes containing bdippza. Thus, the magnesium complex Mg(bdippza) 2 was prepared by an analogous procedure to the preparation of Zn(bdippza) 2 (Scheme 29) In the first effort to prepare a magnesium complex, a solution containing base deprotonated bdippza in methanol was added to a solution of Mg(OTf) 2 in methanol. Scheme 30 : Preparation of Mg(bdippza) 2 from Hbdippza and MgX 2 in basic meth anol, where X = OTf or Cl
84 Unlike the reaction of bdippza with Zn(OTf) 2 addition of the ligand solution to the metal solution did not immediately result in a precipitate. Over the span of several days, however, crystals precipitated without evaporation of methanol being necessary. A proton NMR spectrum and a carbon NMR spectrum both indicated that the crystals were composed of Mg(bdippza) 2 The spectra appeared almost identical to the spectra of Zn(bdippza) 2 as would only be expected if the complexes ha d analogous bis ligand, octahedral structures. The reaction of bdippza with MgCl 2 also resulted in slow precipitation of crystals of Mg(bdippza) 2 rather than a different complex as was the case for the reaction of bdippza with ZnCl 2 The proton NMR spect rum ( Figure 37 and Figure 54 in Appendix A ) showed four doublets between 0.93 and 1.41 ppm, each integrating for twelve m ethyl protons, septets at 2.60 and 3.07 ppm each in tegrating for four isopropyl methine protons, a singlet at 5.88 ppm integrating for four pyrazolyl protons, and a singlet at 6.61 ppm integrating for two central methine protons. As previously mentioned, this NMR spectrum is almost exactly identical to th e proton NMR spectrum of Zn(bdippza) 2 indicating that the compound is surely Mg(bdippza) 2 The carbon NMR spectrum ( Figure 55 ) showed six resonances between 22.0 and 27.2 ppm corresponding to the isopropyl carbon s, a resonance at 68.5 ppm corresponding to the central methine carbon, resonances at 99.1, 151.7, and 161.9 ppm corresponding to the pyrazolyl carbons, and a resonance at 166.1 ppm corresponding to the carboxylate carbon. Once again, the carbon NMR is al most identical to the carbon NMR of Zn(bdippza) 2 The FT IR spectrum of Mg(bdippza) 2 ( Figure 56 ) shows a carbonyl stretch centered at 1666 cm 1 which is slightly higher energy than the carbonyl stretch for Zn(bdi ppza) 2 centered at 1654 cm 1 This indicates a
85 stronger C=O bond in Mg(bdippza) 2 than in Zn(bdippza) 2 due to the metal bound carboxylate oxygen donating more stron gly to magnesium than to zinc. Although good quality crystals of Mg(bdippza) 2 have been gro wn from both methanol and chloroform, a structure has not yet been dete rmined by x ray crystallography. However, the excellent agreement with the 1 H NMR and 13 C NMR spectra of Zn(bdippza) 2 is sufficient to claim that the compound is Mg(bdippza) 2 Figure 37 : 1 H NMR spectrum of Mg(bdippza) 2 in CDCl 3
86 Since reaction of the deprotonated bdippza ligand with magnesium chloride and magnesium triflate both resulted in very slow precipitation of Mg(bdippza) 2 it was unclear what s pecies we re present in the methanol solution. Thus, the reaction of bdippza with one equivalent of Mg(OTf) 2 was carried out in an NMR tube in CD 3 OD with one equivalent of added NaOH and followed by 1 H NMR spectroscopy. The proton NMR spectrum of deprotonated bdi ppza in CD 3 OD ( Figure 38 ) showed two doublets in close proximity at 0.93 and 0.97 ppm, integrating collectively for twelve methyl protons, a third methyl at 1.18 ppm, integrating for twelve methyl protons, two sept ets at 2.87 and 3.26 ppm each integrating for two isopropyl methine protons, a singlet at 5.92 ppm integrating for two pyrazolyl protons, and a singlet at 6.87 ppm integrating for one central methine proton. The peaks were referenced relative to the CD 3 OH peak, which appears at 4.87 ppm. The resonance corresponding to CD 2 HOD appears at 3.28 ppm, and therefore overlaps with one of the isopropyl methine resonances. Upon combining this solution with a CD 3 OD solution containing Mg(OTf) 2 several changes occu r in the proton NMR ( Figure 39 ). All of the peaks corresponding to the bdippza ligand broaden, most significantly the two doublets and the septet corresponding to one of the isopropyl groups T he two doublets hav e broadened into a broad singlet centered at 1.02 ppm, and the septet has broadened into another broad singlet centered at 2.93 ppm. The resonances corresponding to the pyrazolyl protons and the central methine proton have also broadened signif ic antly bu t t he doublet and septet corresponding to the other isopropyl group s have only broadened slightly so that they still appear as a doublet and a septet The broadening likely means that the environments of the protons are fluxional on the NMR timescale p ossibly due to exchange of the ligand between magnesium and sodium
87 centers, or due to fast interconversion between different coordination modes, possibly involving multiple metal centers. The peaks that still appear as a doublet and a septet likely corres pond to the isopropyl group at the 5 position of the pyrazole rings, since they are a greater distance from the nitrogen that binds to the metal; however, a definite assignment cannot be made based on this information. Figure 38 : 1 H NMR of Na(bdippza) in CD 3 OD 1 1 2 3 3 4 4
88 Figure 39 : 1 H NMR of mixture of bdippza and Mg(OTf) 2 in CD 3 OD It is clear that no Mg(bdippza) 2 is present in solution, even after 24 hours, at which point crystals of Mg(bdippza) 2 could be seen to have precipitated on the walls of the NMR tube. The proton NMR spectrum appears essentially identical at this point to the proton NMR spectrum taken immediately after mixing the bdippza solution with the Mg(OTf) 2
89 solution. This means t hat Mg(bdippza) 2 like Zn(bdippza) 2 is very insoluble in methanol, and most likely precipitates out of solution as soon as it is formed. In another effort to prepare a mono ligand sodium complex of bdippza, Hbdippza reacted with di n butylmagnesium in a s prepare compound 3 shown in Scheme 19 60 This procedure does not require the addition of base, since di n b utylmagnesium is a strong base in addition to being a good source of magnesium. The butyl groups are easily displaced by a protic reagent, such as Hbdippza, and leave the reaction vessel as butane gas after being protonated. Due to insolubility of H bdipp za in instead carried out in THF. A powder was isolated after carrying out the reaction and removing solvent. Unfortunately, the proton NMR spectrum ( Figure 40 ) showed that the only bdippza containing product of the reaction was again Mg(bdippza) 2 Peaks corresponding to butyl groups could also be seen in the proton NMR spectrum, overlapping with the four doublets corresponding to the meth yl protons in Mg(bdippza) 2 This indicated the presence of unreacted Mg( n Bu) 2 Clearly, binding of one bdippza ligand activates the resulting complex toward coordination of another bdippza ligand, and bdippza does not have enough steric bulk to make doubl e coordination unfavorable. Thus, it has not, as of yet, been possible to isolate a mono ligand magnesium complex containing bdippza.
90 Figure 40 : 1 H NMR spectrum of Mg(bdippza) 2 and Mg( n Bu) 2 in CDCl 3 Peaks corresponding to th e butyl protons overlap with the methyl peaks of Mg(bdippza) 2 between 0.9 and 1.6 ppm. The peak at 3.65 ppm likely corresponds to THF solvent. Chapter 4 : Conclusions and Future Directions Two zinc compounds and one magnesium compound containing the new m onoanionic [N 2 O] heteroscorpionate ligand bdippza have been isolated. Good quality crystals have been grown for each of these complexes. The structure of Zn(bdippza) 2 has been
91 determined via single crystal x ray diffraction, and crystallographic structur e determination of the other isolated zinc compound is currently pending. Zn(bdippza) 2 and Mg(bdippza) 2 are both distorted octahedral, bis ligand complexes, indicating that bdippza does not have sufficient steric hindrance to make double coordination unfa vorable. [Zn(bdippza)Cl] 2 a zinc complex that contains only one equivalent of bdippza per zinc atom can, however, be isolated in good yield. Hence, zinc complexes containing bdippza may potentially be applied to synthetic modeling of [N 2 O] zinc protease enzymes. Mg(bdippza) 2 is the first isolated magnesium complex containing a bispyrazolylacetate ligand. Although several procedures that showed success for similar systems were attempted in order to isolate a mono ligand magnesium complex containing bdip pza, Mg(bdippza) 2 was the only magnesium compound containing a bdippza ligand that could be isolated. Burzlaff and coworkers have found t hat bdtbpza coordinates only once to zinc, resulting in tetrahedral, mono ligand complexes. 51 Therefore, it is likely that bdtbpza will also coordinate only once to magnesium Hence, a future goal of this project is to prepare tetrahedral magnesium complexes containing bdtbpza. The bulkier bdtbpza ligand is more like ly to support tetrahedral coordination, but a mono ligand bdtbpza magnesium compound is also less likely to support the coordination of a n additional small bidentate ligand such as acetol than an analogous bdippza compound Coordination of a molecule of a cetol, at least in solution, is necessary in order to model the enediolization step catalyzed by RuBisCo. As discussed earlier, Hbdtbpza is prepared via the same route as Hbdippza prepared herein. It is likely that reaction of one equivalent of Hbdtbpza with one equivalent of Mg( n Bu) 2 will result in complete conversion to (bdtbpza)Mg( n Bu),
92 yielding only butane gas as a side product. The n butyl group could then be easily displaced by another ligand. It should be noted that although a mono ligand bispyraz olylacetate magnesium complex would be a reasonable compound for modeling the structure of the active site of RuBisCo, the active site actually contains a facially bound [O 3 ] scheme, rather than an [N 2 O] scheme. Hence, a closer structural approximation to the active site of RuBisCo may result from the use of an [O 3 ] ligand, rather than an [N 2 O] ligand like bdippza. The Parkin group and the Klaui group have each reported facially binding, monoanionic [O 3 ] ligands but neither binds at a carboxylate oxygen, as is desired. 34 63 Therefore, another possible future research direction involves the preparation of new [O 3 ] scorpionate ligands that bind at least one carboxylate oxygen in order to prepare closer synthetic models to the active site of RuBisCo Because magnesium has a much higher affinity for oxygen than nitrogen, an [O 3 ] ligand would likely bind more strongly to a magnesium center than an [N 2 O] li gand. Scheme 31 shows a proposed synthesis for bis 2 oxo 1 tert butylbenzimidazol 1 ylacetic acid (H botbia ). The proposed target can likely be deprotonated and used as a monoanionic [O 3 ] heteroscorpionate ligand. In this proposed synthesis, H botbia is prepared in five steps beginning with the commercially available starting material 1 fluoro 2 nitrobenzene The first three steps have been previously reported in the literature. 64
93 Scheme 31 : Proposed synthesis of H botbia a compound that may act as an [O 3 ] heteroscorpionate ligand in deprotonated form First, 1 fluoro 2 nitrobenzene reacts with tert butylamine in DMF to give tert butyl ( 2 nitrophenyl ) amine via a nucleophilic aromatic substitution. Next, the nitro group is reduced to an amine group to give N tert butyl 1,2 diamine using palladium on carbon as a reduction catalyst, and by generating H 2 in situ from methanol and sodium boro hydride. In the third step, 1 tert butyl benzimidazol 2 one is generated from reaction of N tert butyl 1,2 diamine carbonyldiimidazole In the next step, 2 equivalents of 1 tert butyl benzimidazol 2 one reacts with dichloromethane in the prese nce of base and a phase transfer catalyst to form bis 2 oxo 1 tert butyl benzimidazol 3 ylmethane ( botbim ) This reaction has not, as of yet, been applied to imidazolone compounds, but it
94 has been shown to afford bisazolylmethane compounds from imidazole and benzimidazole precursors under the same conditions as used to afford bispyrazolylmethanes from pyrazole precursors. 38 A possible difficulty in applying this reaction to benzimidizolones is that the o xygens attached to the imidazole rings may be nucleophilic, since the compound has an aromatic resonance form that puts a formal negative charge on the oxygen. Since either the oxygen or the nitrogen could be viable nucleophiles, either may attack dichlor omethane, so that some or all of the product is the ether that results from nucleophilic attack by the oxygen rather than the nitrogen. Reaction conditions may have to be altered in order to successfully prepare Hbotbia but it is likely that bis benzimi dazol 3 ylmethanes can be prepared under solid liquid phase transfer conditions. On ce botbim has been isolated, preparation of H botbia should be relatively straightforward. botbim would likely be readily deprotonated by butyllithium, and would then attac k carbon dioxide to form botbia which can be isolated as H botbia after acid workup. Botbia would likely bind to most metals with an [O 3 ] binding scheme, as is seen in To R ligands ( Figure 15 ). However, instead of binding at three oxoimidazolyl oxygens, it would bind at two oxoimidazolyl oxygens and a carboxylate oxygen. Hence, facial binding of a single molecule of botbia to a magnesium center would result in a closer structural model for RuBisCo than binding of e ither bdippza or To tBu [O 3 ] Binding of botbia to a metal would result in two seven membered chelate rings and an eight membered chelate ring so that the botbia may have somewhat more flexibility than bispyrazolylacetates, which contain three six membere d chelate rings. However, six membered rings are often intrinsically more thermodynamically stable that larger chelate
95 rings, and thus botbia may bind more weakly than bdippza in some cases. When preparing magnesium complexes, however, [O 3 ] binding to ma gnesium rather than [N 2 O] binding has the opposite effect, strengthening the binding of botbia relative to bdippza, since the oxygen donor atoms are harder than the nitrogen donor atoms. A potential problem in using complexe s containing botbia to model e nz ymes is that despite the considerable steric hindrance imposed by the tert butyl groups, bis ligand complexes may still be favored. The tert butyl group is more distant from the metal center than in analogous bdtbpza complexes, and there is only one ter t butyl substituent rather than two. Hence, botbia may not have enough hindrance to favor mono ligand complexes over bis ligand complexes. If only a bis ligand magnesium complex can be isolated using botbia bis benzimidazolylacetate ligands can be synth esized from benzimidazole precursors containing more hindered groups than tert butyl, such as adamantyl or triisopropylphenyl, substituted at the 1 position of the imidazole ring. Thus, a new class of monoanionic [O 3 ] heteroscorpionate ligands could poten tially be accessed via the route outlined in Scheme 31 for use in applications including, but certainly not limited to, synthetic enzyme modeling.
96 Appendices A.1: Appendix A: Selected NMR and FT IR Spectra Figure 41 : 1 H NMR spectrum of 3,5 diisopropylpyrazole in CDCl 3
97 Figure 42 : 13 C NMR spectrum of 3,5 diisopropylpyrazole in CDCl 3
98 Figure 43 : 1 H NMR spectrum of bdippzm in CDCl 3
99 Figure 44 : 13 C NMR spectrum of bdippzm in CDCl 3
100 Figure 45 : 1 H NMR spectrum of Hbdippza in CDCl 3
101 Figure 46 : 13 C NMR spectrum of Hbdippza in CDCl 3
102 Figure 47 : Solid state FT IR spectrum of Hbdippza
103 Figure 48 : 1 H NMR spectrum of Zn(bdippza) 2 in CDCl 3
104 Figure 49 : 13 C NMR spectrum of Zn(bdippza) 2 in CDCl 3
105 Figure 50 : Solid state FT IR spectrum of Zn(bd ippza) 2
106 Figure 51 : 1 H NMR spectrum of [Zn(bdippza)Cl] 2 in CDCl 3
107 Figure 52 : 13 C NMR spectrum of [Zn(bdippza)Cl] 2 in CDCl 3
108 Figure 53 : Solid state FT IR spectrum of [Zn(bdippza)C l] 2
109 Figure 54 : 1 H NMR spectrum of Mg(bdippza) 2 in CDCl 3
110 Figure 55 : 13 C NMR spectrum of Mg(bdippza) 2 in CDCl 3
111 Figure 56 : Solid state FT IR spectrum of Mg(bdippza) 2
112 A.2: Append ix B: Crystallographic Information: Written by Dr. Lee Daniels A.2.1 Crystallographic information for Zn(bdippza) 2 Experimental Data Collection A colorless prism crystal of C 40 H 60 N 8 O 4 Zn having approximate dimensions of 0.200 x 0.190 x 0.180 mm was mounted on a glass fiber. All measurements were made on a Rigaku XtaLAB mini diffractometer using graphite monochromated Mo K radiation. The crystal to detector distance was 48.80 mm. Cell constants and an orientation matrix for data collection corresponded to a primitive monoclinic cell with d imensions: a = 10.166(4) b = 17.284(6) = 96.866(7) o c = 12.528(4) V = 2185(2) 3
113 For Z = 2 and F.W. = 782.35, the calculated density is 1.189 g/cm 3 The reflection conditions of: h0l: l = 2n 0k0: k = 2n uniquely determine the space group to be: P2 1 /c (#14) The data were collected at a temperature of 20 + 1 o 55.0 o A total of 360 oscillation images were collected. A sweep of da = 0 o using w oscillations from 60.0 to 120.0 o in 1.0 o steps. A second sweep was o using w oscillations from 60.0 to 120.0 o in 1.0 o steps. The exposure rate was 30.0 [sec./ o ]. The detector swing angle was 29.90 o The crystal to detector distance was 48.80 mm. Readout was performed in the 0.146 mm pixel mode. Data Reduction Of the 15227 reflections that were collected, 5002 were unique (R int = 0.0318); equivalent reflections were merged. Data were collected and processed using CrystalClear (Rigaku).
114 The linear absorption coefficient, for Mo K radiation is 6.082 cm 1 An empirical absorption correction was applied which resulted in transmission factors ranging from 0.723 to 0.896. The data were corrected for Lorentz and polarization effects. St ructure Solution and Refinement The structure was solved by direct methods 2 and expanded using Fourier techniques. The non hydrogen atoms were refined anisotropically. Hydrogen atoms were refined using the riding model. The final cycle of full matrix lea st squares refinement 3 on F 2 was based on 4997 observed reflections and 249 variable parameters and converged (largest parameter shift was 0.00 times its esd) with unweighted and weighted agreement factors of: R1 = ||Fo| |Fc|| / |Fo| = 0.0398 wR2 = [ ( w (Fo 2 Fc 2 ) 2 )/ w(Fo 2 ) 2 ] 1/2 = 0.0991 The standard deviation of an observation of unit weight 4 was 1.02. Unit weights were used. The maximum and minimum peaks on the final difference Fourier map corresponded to 0.23 and 0.26 e / 3 respectively. Neutral atom scattering factors were taken from Cromer and Waber 5 Anomalous dispersion effects were included in Fcalc 6 ; the values for f' and f" were those of Creagh and McAuley 7 The values for th e mass attenuation coefficients are those of Creagh and Hubbell 8 All calculations were performed using the CrystalStructure 9
115 crystallographic software package except for refinement, which was performed using SHELXL 97 10
116 References (1) CrystalClear : Rigaku Corporation, 1999. CrystalClear Software User's Guide, Molecular Structure Corporation, (c) 2000.J.W.Pflugrath (1999) Acta Cryst. D55, 1718 1725. (2) SHELX97 : Sheldrick, G.M. (2008). Acta Cryst. A64, 112 122. (3) Least Squares function minimized: (SHELXL97) w (F o 2 F c 2 ) 2 where w = Least Squares weights. (4) Standard deviation of an observation of unit weight: [ w (F o 2 F c 2 ) 2 /(N o N v )] 1/2 where: N o = number of observations N v = number of variables (5) Cromer, D. T. & Wa ber, J. T.; "International Tables for X ray Crystallography", Vol. IV, The Kynoch Press, Birmingham, England, Table 2.2 A (1974). (6) Ibers, J. A. & Hamilton, W. C.; Acta Crystallogr., 17, 781 (1964).
117 (7) Creagh, D. C. & McAuley, W.J .; "International Tables for Crystallography", Vol C, (A.J.C. Wilson, ed.), Kluwer Academic Publishers, Boston, Table 188.8.131.52, pages 219 222 (1992). (8) Creagh, D. C. & Hubbell, J.H..; "International Tables for Crystallography", Vol C, (A.J.C. Wilson, ed.), Kluwer Academi c Publishers, Boston, Table 184.108.40.206, pages 200 206 (1992). (9) CrystalStructure 4.0 : Crystal Structure Analysis Package, Rigaku Corporation (2000 2010). Tokyo 196 8666, Japan. (10) SHELX97 : Sheldrick, G.M. (2008). Acta Cryst. A64, 112 122.
118 EXPERIMENTA L DETAILS A. Crystal Data Empirical Formula C 40 H 60 N 8 O 4 Zn Formula Weight 782.35 Crystal Color, Habit colorless, prism Crystal Dimensions 0.200 X 0.190 X 0.180 mm Crystal System monoclinic Lattice Type Primitive Lattice Parameters a = 10.166(4 ) b = 17.284(6) c = 12.528(4) = 96.866(7) o V = 2185(2) 3
119 Space Group P2 1 /c (#14) Z value 2 D calc 1.189 g/cm 3 F 000 836.00 (MoK ) 6.082 cm 1
120 B. Intensity Measurements Diffractometer XtaLAB mini Radiation MoK ( = 0.71075 ) gra phite monochromated Voltage, Current 50kV, 12mA Temperature 20.0 o C Detector Aperture 75 mm (diameter) Data Images 360 exposures o ) 60.0 120.0 o o ) 60.0 120.0 o Exposure Rate 30.0 sec./ o Det ector Swing Angle 29.90 o
121 Detector Position 48.80 mm Pixel Size 0.146 mm 2 max 55.0 o No. of Reflections Measured Total: 15227 Unique: 4997 (R int = 0.0318) Corrections Lorentz polarization Absorption (trans. factors: 0.723 0.896)
122 C. Structure S olution and Refinement Structure Solution Direct Methods (SHELX97) Refinement Full matrix least squares on F 2 Function Minimized w (Fo 2 Fc 2 ) 2 Least Squares Weights w = 1/ [ 2 (Fo 2 ) + (0.0422 P) 2 + 0.6940 P ] where P = (Max(Fo 2 ,0) + 2F c 2 )/3 2 max cutoff 55.0 o Anomalous Dispersion All non hydrogen atoms No. Observations (All reflections) 4997 No. Variables 249 Reflection/Parameter Ratio 20.07
123 Residuals: R1 (I>2.00 (I)) 0.0398 Residuals: R (All reflections) 0.0600 Residuals: wR2 ( All reflections) 0.0991 Goodness of Fit Indicator 1.016 Max Shift/Error in Final Cycle < 0.001 Maximum peak in Final Diff. Map 0.23 e / 3 Minimum peak in Final Diff. Map 0.26 e / 3
124 Table 1. Atomic coordinates and B iso /B eq and occupancy atom x y z B eq occ Zn(1) 0.5000 0.0000 0.0000 3.002(8) 1/2 O(1) 0.3285(2) 0.05637(8) 0.0517(1) 3.53(3) 1 O(2) 0.1103(2) 0.0481(1) 0.0809(2) 4.50(4) 1 N(1) 0.3852(2) 0.10666(9) 0.0232(2) 3.03(3) 1 N(2) 0.2663(2) 0.10551(9 ) 0.0178(2) 2.81(3) 1 N(3) 0.2957(2) 0.00786(9) 0.1538(2) 2.99(3) 1 N(4) 0.4296(2) 0.00352(9) 0.1568(2) 3.08(3) 1 C(1) 0.2173(2) 0.0325(1) 0.0541(2) 2.69(3) 1 C(2) 0.2179(2) 0.0299(1) 0.0349(2) 2.80(3) 1 C(3) 0.6194(3 ) 0.2461(2) 0.0083(3) 6.97(8) 1 C(4) 0.5274(3) 0.2052(2) 0.0935(2) 4.92(5) 1 C(5) 0.4952(4) 0.2562(2) 0.1915(3) 7.60(9) 1 C(6) 0.4043(2) 0.1798(1) 0.0484(2) 3.56(4) 1 C(7) 0.2984(2) 0.2250(2) 0.0249(2) 4.16(5) 1 C(8) 0.2127(2) 0.1772(1) 0.0182(2) 3.38(4) 1 C(9) 0.0193(3) 0.2657(2) 0.0106(3) 6.82(8) 1 C(10) 0.0854(2) 0.1941(2) 0.0628(2) 4.41(5) 1 C(11) 0.1093(3) 0.2036(2) 0.1839(3) 7.05(8) 1 C(12) 0.0573(3) 0.0963(2) 0.2202(3) 8.07(9) 1 C(13) 0.1123(3) 0.0196(2) 0.2650(2) 4.90(6) 1 C(14) 0.0941(3) 0.0114(3) 0.3825(3) 8.9(2) 1
125 C(15) 0.2551(2) 0.0121(2) 0.2486(2) 3.63(4) 1 C(16) 0.3686(2) 0.0289(2) 0.3156(2) 4.26(5) 1 C(17) 0.4744(2) 0.0179(1) 0.2561(2) 3. 41(4) 1 C(18) 0.6485(3) 0.0825(2) 0.3826(3) 6.36(7) 1 C(19) 0.6200(2) 0.0256(2) 0.2908(2) 4.01(5) 1 C(20) 0.6796(3) 0.0526(2) 0.3211(3) 5.69(6) 1 B eq = 8/3 2 (U 11 (aa*) 2 + U 22 (bb*) 2 + U 33 (cc*) 2 + 2U 12 (aa*bb*)cos + 2U 13 (aa*cc*)cos + 2U 23 (bb*cc*)cos )
126 Table 2. Atomic coordinates and B iso involving hydrogen atoms atom x y z B iso occ H(1) 0.1256 0.0401 0.0687 3.23 1 H(3A) 0.6424 0.2118 0.0512 8.36 1 H(3B) 0.5760 0.2910 0.0158 8.36 1 H(3C) 0.6983 0.2615 0.0379 8.36 1 H(4) 0.5728 0.1590 0.1156 5.90 1 H(5A) 0.4568 0.3038 0.1706 9.13 1 H(5B) 0.4333 0.2300 0.2431 9.13 1 H(5C) 0.5749 0.2671 0.2228 9.13 1 H(7) 0.2882 0.2779 0.0365 4.99 1 H(9A) 0.0024 0.2580 0 .0657 8.18 1 H(9B) 0.0765 0.3095 0.0254 8.18 1 H(9C) 0.0629 0.2748 0.0391 8.18 1 H(10) 0.0257 0.1500 0.0466 5.29 1 H(11A) 0.1627 0.2487 0.2013 8.46 1 H(11B) 0.1543 0.1588 0.2153 8.46 1 H(11C) 0.0259 0.2094 0.2118 8.4 6 1 H(12A) 0.1063 0.1380 0.2566 9.68 1 H(12B) 0.0653 0.0988 0.1447 9.68 1 H(12C) 0.0343 0.1005 0.2311 9.68 1 H(13) 0.0631 0.0222 0.2252 5.88 1 H(14A) 0.1225 0.0392 0.4073 10.73 1
127 H(14B) 0.1459 0.0499 0.4236 10.73 1 H( 14C) 0.0022 0.0181 0.3912 10.73 1 H(16) 0.3738 0.0447 0.3870 5.11 1 H(18A) 0.6098 0.1317 0.3617 7.63 1 H(18B) 0.6112 0.0635 0.4445 7.63 1 H(18C) 0.7425 0.0883 0.3998 7.63 1 H(19) 0.6617 0.0452 0.2296 4.82 1 H(20A) 0.6 382 0.0737 0.3797 6.83 1 H(20B) 0.6655 0.0868 0.2605 6.83 1 H(20C) 0.7730 0.0470 0.3426 6.83 1
128 Table 3. Anisotropic displacement parameters atom U 11 U 22 U 33 U 12 U 13 U 23 Zn(1) 0.0245(2) 0.0493(2) 0.0404(2) 0.0007(2) 0.004 3(1) 0.0026(2) O(1) 0.0292(7) 0.0541(9) 0.0507(8) 0.0007(6) 0.0040(6) 0.0129(7) O(2) 0.0301(7) 0.077(1) 0.062(1) 0.0074(7) 0.0009(7) 0.0228(8) N(1) 0.0272(8) 0.0434(9) 0.0454(9) 0.0011(7) 0.0085(7) 0.0005(7) N(2) 0.0262(7) 0.0421(9) 0.038 9(9) 0.0002(7) 0.0052(6) 0.0012(7) N(3) 0.0268(7) 0.054(1) 0.0331(8) 0.0005(7) 0.0034(6) 0.0011(7) N(4) 0.0265(7) 0.053(1) 0.0372(8) 0.0031(7) 0.0012(6) 0.0010(8) C(1) 0.0227(8) 0.046(1) 0.0331(9) 0.0007(8) 0.0026(7) 0.0021(8) C(2) 0.0 292(9) 0.042(1) 0.035(1) 0.0037(8) 0.0035(7) 0.0029(8) C(3) 0.054(2) 0.097(3) 0.114(3) 0.030(2) 0.007(2) 0.003(2) C(4) 0.056(2) 0.050(2) 0.086(2) 0.011(1) 0.030(2) 0.001(2) C(5) 0.109(3) 0.101(3) 0.082(2) 0.043(2) 0.026(2) 0.013(2) C(6) 0.040(1) 0.043(2) 0.052(2) 0.0063(9) 0.0063(9) 0.0030(9) C(7) 0.046(2) 0.035(1) 0.078(2) 0.0026(9) 0.012(1) 0.005(1) C(8) 0.034(1) 0.044(2) 0.050(2) 0.0004(9) 0.0024(9) 0.0108(9) C(9) 0.053(2) 0.072(2) 0.134(3) 0.020(2) 0.009(2) 0.010( 2) C(10) 0.038(1) 0.050(2) 0.081(2) 0.001(1) 0.015(1) 0.017(2) C(11) 0.083(2) 0.106(3) 0.085(2) 0.012(2) 0.032(2) 0.027(2) C(12) 0.082(2) 0.124(3) 0.101(3) 0.046(2) 0.016(2) 0.013(2) C(13) 0.043(2) 0.107(2) 0.037(1) 0.009(2) 0.0095(9) 0.007 (2) C(14) 0.058(2) 0.239(5) 0.046(2) 0.000(2) 0.020(2) 0.004(2)
129 C(15) 0.040(1) 0.065(2) 0.033(1) 0.002(1) 0.0058(8) 0.0012(9) C(16) 0.049(2) 0.080(2) 0.032(1) 0.003(1) 0.0002(9) 0.006(1) C(17) 0.041(1) 0.050(2) 0.036(1) 0.0053(9) 0.0031(8) 0.0048(8) C(18) 0.074(2) 0.080(2) 0.079(2) 0.016(2) 0.027(2) 0.010(2) C(19) 0.040(1) 0.066(2) 0.044(2) 0.012(1) 0.0084(9) 0.008(1) C(20) 0.051(2) 0.076(2) 0.082(2) 0.000(2) 0.019(2) 0.004(2) The general temperature factor expression: exp( 2 2 (a* 2 U 11 h 2 + b* 2 U 22 k 2 + c* 2 U 33 l 2 + 2a*b*U 12 hk + 2a*c*U 13 hl + 2b*c*U 23 kl))
130 Table 4. Bond lengths () atom atom distance atom atom distance Zn(1) O(1) 2.0342(14) Zn(1) O(1) 1 2.0342(14) Zn(1) N(1) 2.1827(17) Zn(1) N(1) 1 2.1827(17) Zn(1) N(4) 2.1704(1 7) Zn(1) N(4) 1 2.1704(17) O(1) C(2) 1.256(3) O(2) C(2) 1.216(3) N(1) N(2) 1.368(3) N(1) C(6) 1.322(3) N(2) C(1) 1.450(3) N(2) C(8) 1.353(3) N(3) N(4) 1.360(2) N(3) C(1) 1.463(3) N(3) C(15) 1.348(3) N(4) C(17) 1.326(3) C(1) C(2) 1.550(3) C(3) C (4) 1.507(4) C(4) C(5) 1.515(4) C(4) C(6) 1.500(4) C(6) C(7) 1.389(3) C(7) C(8) 1.359(3) C(8) C(10) 1.498(3) C(9) C(10) 1.519(4) C(10) C(11) 1.516(4) C(12) C(13) 1.520(5) C(13) C(14) 1.512(4) C(13) C(15) 1.496(4) C(15) C(16) 1.373(3) C(16) C(17 ) 1.393(3) C(17) C(19) 1.498(3) C(18) C(19) 1.515(4) C(19) C(20) 1.510(4) Symmetry Operators: (1) X+1, Y, Z
13 1 Table 5. Bond lengths involving hydrogens () atom atom distance atom atom distance C(1) H(1) 0.980 C(3) H(3A) 0.960 C(3) H(3B) 0.9 60 C(3) H(3C) 0.960 C(4) H(4) 0.980 C(5) H(5A) 0.960 C(5) H(5B) 0.960 C(5) H(5C) 0.960 C(7) H(7) 0.930 C(9) H(9A) 0.960 C(9) H(9B) 0.960 C(9) H(9C) 0.960 C(10) H(10) 0.980 C(11) H(11A) 0.960 C(11) H(11B) 0.960 C(11) H(11C) 0.960 C(12) H(12A) 0.960 C(12) H(12B) 0.960 C(12) H(12C) 0.960 C(13) H(13) 0.980 C(14) H(14A) 0.960 C(14) H(14B) 0.960 C(14) H(14C) 0.960 C(16) H(16) 0.930 C(18) H(18A) 0.960 C(18) H(18B) 0.960 C(18) H(18C) 0.960 C(19) H(19) 0.980 C(20) H(20A) 0.960 C(20) H(20B) 0.960 C(20) H(20C) 0.960
132 Table 6. Bond angles ( o ) atom atom atom angle atom atom atom angle O(1) Zn(1) O(1) 1 180.00(8) O(1) Zn(1) N(1) 86.44(6) O(1) Zn(1) N(1) 1 93.56(6) O(1) Zn(1) N(4) 86.50(6) O(1) Zn(1) N(4) 1 93.50(6) O(1) 1 Zn(1) N(1) 93.56 (6) O(1) 1 Zn(1) N(1) 1 86.44(6) O(1) 1 Zn(1) N(4) 93.50(6) O(1) 1 Zn(1) N(4) 1 86.50(6) N(1) Zn(1) N(1) 1 180.00(8) N(1) Zn(1) N(4) 82.40(6) N(1) Zn(1) N(4) 1 97.60(6) N(1) 1 Zn(1) N(4) 97.60(6) N(1) 1 Zn(1) N(4) 1 82.40(6) N(4) Zn(1) N(4) 1 180.00(8) Zn(1) O(1 ) C(2) 121.51(13) Zn(1) N(1) N(2) 114.84(12) Zn(1) N(1) C(6) 138.37(13) N(2) N(1) C(6) 105.36(15) N(1) N(2) C(1) 118.84(15) N(1) N(2) C(8) 111.25(15) C(1) N(2) C(8) 129.91(16) N(4) N(3) C(1) 118.76(15) N(4) N(3) C(15) 111.85(15) C(1) N(3) C(15) 129.39( 16) Zn(1) N(4) N(3) 114.46(11) Zn(1) N(4) C(17) 136.75(13) N(3) N(4) C(17) 105.77(15) N(2) C(1) N(3) 110.38(14) N(2) C(1) C(2) 110.42(15) N(3) C(1) C(2) 111.31(15) O(1) C(2) O(2) 126.79(19) O(1) C(2) C(1) 117.04(15) O(2) C(2) C(1) 116.17(17) C(3) C(4) C(5) 110.9(3) C(3) C(4) C(6) 110.3(3) C(5) C(4) C(6) 111.5(2) N(1) C(6) C(4) 121.56(18) N(1) C(6) C(7) 110.32(18) C(4) C(6) C(7) 128.11(19) C(6) C(7) C(8) 106.98(19) N(2) C(8) C(7) 106.08(18)
133 N(2) C(8) C(10) 123.13(18) C(7) C(8) C(10) 130.77(19) C(8) C(10) C(9) 110.9(2) C(8) C(10) C(11) 110.62(19) C(9) C(10) C(11) 110.6(3) C(12) C(13) C(14) 111.0(3) C(12) C(13) C(15) 110.1(3) C(14) C(13) C(15) 111.05(19) N(3) C(15) C(13) 123.16(17) N(3) C(15) C(16) 105.63(18) C(13) C(15) C(16) 130.99(19) C(15) C( 16) C(17) 106.87(18) N(4) C(17) C(16) 109.85(17) N(4) C(17) C(19) 120.90(18) C(16) C(17) C(19) 129.24(18) C(17) C(19) C(18) 111.71(19) C(17) C(19) C(20) 110.26(18) C(18) C(19) C(20) 110.8(2) Symmetry Operators: (1) X+1, Y, Z
134 Table 7. Bond angles involving hydrogens ( o ) atom atom atom angle atom atom atom angle N(2) C(1) H(1) 108.2 N(3) C(1) H(1) 108.2 C(2) C(1) H(1) 108.2 C(4) C(3) H(3A) 109.5 C(4) C(3) H(3B) 109.5 C(4) C(3) H(3C) 109.5 H(3A) C(3) H(3B) 109.5 H(3A) C(3) H(3C) 109.5 H(3B) C( 3) H(3C) 109.5 C(3) C(4) H(4) 108.0 C(5) C(4) H(4) 108.0 C(6) C(4) H(4) 108.0 C(4) C(5) H(5A) 109.5 C(4) C(5) H(5B) 109.5 C(4) C(5) H(5C) 109.5 H(5A) C(5) H(5B) 109.5 H(5A) C(5) H(5C) 109.5 H(5B) C(5) H(5C) 109.5 C(6) C(7) H(7) 126.5 C(8) C(7) H(7) 1 26.5 C(10) C(9) H(9A) 109.5 C(10) C(9) H(9B) 109.5 C(10) C(9) H(9C) 109.5 H(9A) C(9) H(9B) 109.5 H(9A) C(9) H(9C) 109.5 H(9B) C(9) H(9C) 109.5 C(8) C(10) H(10) 108.2 C(9) C(10) H(10) 108.2 C(11) C(10) H(10) 108.2 C(10) C(11) H(11A) 109.5 C(10) C(11) H (11B) 109.5 C(10) C(11) H(11C) 109.5 H(11A) C(11) H(11B) 109.5 H(11A) C(11) H(11C) 109.5 H(11B) C(11) H(11C) 109.5 C(13) C(12) H(12A) 109.5 C(13) C(12) H(12B) 109.5 C(13) C(12) H(12C) 109.5 H(12A) C(12) H(12B) 109.5 H(12A) C(12) H(12C) 109.5 H(12B) C( 12) H(12C) 109.5 C(12) C(13) H(13) 108.2
135 C(14) C(13) H(13) 108.2 C(15) C(13) H(13) 108.2 C(13) C(14) H(14A) 109.5 C(13) C(14) H(14B) 109.5 C(13) C(14) H(14C) 109.5 H(14A) C(14) H(14B) 109.5 H(14A) C(14) H(14C) 109.5 H(14B) C(14) H(14C) 109.5 C(15) C(1 6) H(16) 126.6 C(17) C(16) H(16) 126.6 C(19) C(18) H(18A) 109.5 C(19) C(18) H(18B) 109.5 C(19) C(18) H(18C) 109.5 H(18A) C(18) H(18B) 109.5 H(18A) C(18) H(18C) 109.5 H(18B) C(18) H(18C) 109.5 C(17) C(19) H(19) 108.0 C(18) C(19) H(19) 108.0 C(20) C(19) H(19) 108.0 C(19) C(20) H(20A) 109.5 C(19) C(20) H(20B) 109.5 C(19) C(20) H(20C) 109.5 H(20A) C(20) H(20B) 109.5 H(20A) C(20) H(20C) 109.5 H(20B) C(20) H(20C) 109.5
136 Works Cited 1. Trofimenko, S. J. Amer. Chem. Soc. 1967, 89, 3170 3177. 2. Trofimenko, S. Chem. Rev. 1993, 93, 943 980. 3. Hegelmann, I.; Beck, A.; Eichorn, C.; Weibert, B.; Burzlaff, N. Eur. J. Inorg. Chem. 2003, 339 347. 4. Otero, A.; Fernandez Baeza, J.; Antinolo, A.; Tejeda, J.; Lara Sanchez, A.; Sanchez Barba, L.; Exposito, M. T.; Rodriguez, A. M. Dalton Trans. 2003, 1614 1619. 5. Otero, A.; Fernandez Baeza, J.; Lara Sanchez, A.; Tejeda, J.; Sanchez Barba, L. Eur. J. Inorg. Chem. 2008, 5309 5326. 6. Pettinari, C. Scorpionates II: Chelating Borate Ligands; Imperial College Press, 2008. 7. Long, G. J.; Hutchinson, B. B. Inorg. Chem 1987, 26, 608 613. 8. Tolman, W. B. Inorg. Chem. 1991, 30, 4877 4880. 9. Trofimenko, S.; Calabrese, J. C.; Thompson, J. S. Inorg. Chem. 1987, 1508 1514. 10. Calabrese, J. C.; Domaille, P. J. T. J. S.; Trofimenko, S. Inorg. Chem. 1990, 29, 4429 4437. 11. Olson, M. D.; Rettig, S. J.; Storr, A.; Trotter, J.; Trofimenko, S. Acta Crystallogr. 1991, C47, 1544 1546. 12. Kitajima, N.; Fukui, H.; Moro oka, Y. J. Am. Chem. Soc. 1990, 112, 6402 6403. 13. Alsfasser, R.; Trofimenko, S.; Looney, A.; Parkin, G.; Vahrenkamp, H. Inorg. Chem. 1991, 30, 4098 4100. 14. Ruf, M.; Weis, K.; Vahrenkamp, H. Inorg. Chem. 1997, 36, 2130 2137. 15. Kitajima, N.; Fujisawa, K.; Fujimoto, C.; Moro oka, Y.; Hashimoto, S.; Kitagawa, T.; Toriumi, K.; Tatsumi, K.; Nakamura, A. J. Am. Chem. Soc. 1992, 114, 1277 1291. 16. Scarrow, R. C.; Maroney, M. J.; Palmer, S. M.; Que, L. J.; Salowe, S. P.; Stubbe, J. A. J. Am. Chem. Soc. 1986, 108, 6832 6834. 17. Parkin, G. Chem. Rev. 2004, 104, 699 767. 18. Chaudhuri, P.; Stockheim, C.; Wieghardt, K.; Deck, W.; Gregorzik, R.; Vahrenkamp, H.; Nuber,
137 B.; Weiss, J. Inorg. Chem. 1992, 31, 1451 1457. 19. Zvargulis, E. S.; Buys, I. E.; Hambley, T. W. Polyhedron 1995, 14, 2267 2273. 20. Lu, D. S.; Voth, G. A. J. Am. Chem. Soc. 1998, 120, 4006 4014. 21. Merz, K. M.; Banci, L. J. Am. Chem. Soc. 1997, 4, 355 365. 22. Looney, A.; Han, R.; McNeill, K.; Parkin, G. J. Am. Chem. Soc. 1993, 115, 4690 4697. 23. Alsfasser, R.; Ruf, M.; Trofimenko, S.; Vahrenkamp, H. Chem. Ber. 1993, 126, 703 710. 24. Brandsch, T.; Schell, F. A.; Weis, K.; Ruf, M.; Mu¨ ller, B.; Vahrenkamp, H. Chem. Ber./Recl. 1997, 130, 283 289. 25. Bergquist, C.; Parkin, G. J. Am. Che m. Soc. 1999, 121, 6322 6323. 26. Garner, M.; Reglinski, J.; Cassidy, I.; Spicer, M. D.; Kennedy, A. R. Chem. Commun. 1996, 1975 1976. 27. Spicer, M. D.; Reglinski, J. Eur. J. Inorg. Chem. 2009, 1553 1574. 28. Garner, M.; Lehmann, M. A.; Reglinski, J.; Spicer, M. D. Organometallics 2001, 20, 5233 5236. 29. Parkin, G. New J. Chem. 2007, 31, 1996 2014. 30. Melnick, J. G.; Parkin, G. Science 2007, 317, 225 227. 31. Bridgewater, B. M.; Fillebeen, T.; Friesner, R. A.; Parkin, G. J. Chem. Soc., Dalton Trans. 2000, 100, 4494 4496. 32. Kimblin, C.; Hascall, T.; Parkin, G. Inorg. Chem. 1997, 36, 5680 5681. 33. Minoura, M.; Landry, V. K.; Melnick, J. G.; Pang, K.; Marchio, L.; Parkin, G. Chem. Commun. 2006, 3990 3992. 34. Al Harbi, A.; Sattler, W.; Sattler, A.; Parkin, G. Chem. Commun. 2011, 47, 3123 3125. 35. Trofimenko, S. J. Am. Chem. Soc. 1970, 92, 5118. 36. Pettinaru, C.; Pettinari, R. Coord. Chem. Rev. 2004, 249, 526 543. 37. Reger, D. L.; Collins, J. E. Inorg. Chem. 1999, 38, 3235 3237.
138 38. Julia, S.; Sala, P.; del Mazo, J.; Sancho, M.; Ochoa, C.; Elguero, J.; Fayet, J. P.; Vertut, M. C. J. Heterocyclic Chem. 1982, 19, 1141 1145. 39. Julia, S.; del Mazo, J.; Avila, L.; Elguero, J. Org. Prep. Proced. Int. 1984, 16 40. Hill, A. F.; Owen, G. R.; White, A. J. P.; Williams, D. J. Angew. Chem., Int. Ed. 1999, 38, 2759 2761. 41. Reger, D. L.; Wright, T. D.; Semeniuc, R. F.; Grattan, T. C.; Smith, M. D. Inorg. Chem. 2001, 40, 6212 6219. 42. Klaui, W.; Berghahn, M.; R heinwald, G.; Lang, H. Angew. Chem., Int. Ed. 2000, 39 (14), 2464 2466. 43. Klaui, W.; Berghahn, M.; Frank, W.; Reib, G. J.; Schonherr, T.; Rheinwald, G.; Lang, H. Eur. J. Inorg. Chem. 2003, 2059 2070. 44. Burzlaff, N. Adv. Inorg. Chem. 2008, 60, 101 162. 45. Dowling, C.; Murphy, V. J.; Parkin, G. Inorg. Chem. 1996, 35, 2415 2420. 46. Dowling, C.; Parkin, G. Polyhedron 1996, 15, 2463 2465. 47. Ghosh, P.; Parkin, G. J. Chem. Soc., Dalton Trans. 1998, 2281 2283. 48. Hammes, B. S.; Carrano, C. J. Inorg. Chem. 1999, 38, 4593 4600. 49. Otero, A.; Fernandez Baeza, J.; Tejeda, J.; Antinolo, A.; Carrillo Hermosilla, F.; Diez Barra, E.; Lara Sanchez, A.; Fernandez Lopez, M.; Lanfranchi, M.; Pellinghelli, M. A. J. Chem. Soc. Dalton Trans. 1999, 3537 3539. 50. Pettinari, C.; Pettinari, R. Coord. Chem. Rev. 2005, 249, 663 691. 51. Beck, A.; Weibert, B.; Burzlaff, N. Eur. J. Inorg. Chem. 2001, 521 527. 52. Hammes, B. S.; Kieber Emmons, M. T.; Letizia, J. A.; Shirin, Z.; Carrano, C. J.; Zakharov, L. N.; Rheingold, A. L. Inorg. Chim. Acta 2003, 346, 227 238. 53. Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L. J. Chem. Rev. 2004, 104, 939 986. 54. S olomon, E. I.; Brunold, T. C.; Davis, M. I.; Kemsley, J. N.; Lee, S. K.; Lehnert, N.; Neese, F.; Skulan, A. J.; Yang, Y. S.; Zhou, J. Chem. Rev. 2000, 100, 235 349.
139 55. Beck, A.; Barth, A.; Hubner, E.; Burzlaff, N. Inorg. Chem. 2003, 42, 7182 7188. 56. Cleland, W. W.; Andrews, T. J.; Gutteridge, S.; Hartman, F. C.; Lorimer, G. H. Chem. Rev. 1998, 98, 549 561. 57. Odom, D.; Gramer, C. J.; Young, V. G. J.; Hilderbrand, S. A.; Sherman, S. E. Inorg. Chim. Acta 2000, 297, 404 410. 58. Hilderb rand, S.; Sherman, S. Unpublished results. 59. Wheeler, K.; Horowitz, J.; Liang, A.; Sherman, S. Unpublished results. 60. Schofield, A. D.; Barros, M. L.; Cushion, M. G.; Schwartz, A. D.; Mountford, P. Dalton Trans. 2009, 85 96. 61. Kitajima, N.; Fujisawa, K.; Fujimoto, C.; Morooka, Y.; Hashimoto, S.; Kitagawa, T.; Toriumi, K.; Tatsumi, K.; Nakamura, A. J. Am. Chem. Soc. 1992, 114, 1277 1291. 62. Otero, A.; Fernandez Baeza, J.; Antinolo, A.; Tejeda, J.; Lara Sanchez, A.; Sanchez Barba, L.; Exposito, M. T.; Rodriguez, A. M. Dalton Trans. 2003, 1614 1619. 63. Klaui, W.; Otto, H.; Eberspach, W.; Bucholz, E. Chem. Ber. 1982, 115, 1922 1933. 64 Zhang, P. e. a. J. Med. Chem. 2009, 52, 5703 5711.